Vous êtes sur la page 1sur 352

TG 1164 JULY 1971 Copy No.

Technical Memorandum

THE MOTION OF BALLISTIC MISSILES


by L. S. GLOVER and J. C. HAGAN

THE JOHNS HOPKINS UNIVERS!TY i APPLIED PHYSICS LABORATORY


Approved for public release; dltrlbutlion uilimited.

NATIONAL TECHNICAL SERVICE INFORMATION Vn 72151


Srnlitoll~d,

R1 1 roduk(d by

,o-

'

THIS DCUMENT iS BEST QUALI Y AVAILABLE. T"I COPY FURN1SHED TO DTIC CONTAINED
A SIGNIFICANT E ICH NUMBER DO OF NOT

REPRODUCED FROM BEST AVAILABLE COPY

t'CLItftt

UNCLASSIFIED I.'lii,.,ifICai ufl DOCUMENT CONTROL DATA - R & D

0 - l;,NlA

'I Nt IF

v ,IT IT,

'C't,0[-td

If. I Mot,)

!0,

P I- POP I

S-_ C J[NlIT

("L A* S

I I.C,

01

The Johns Hopkins University Applied Physics Lab. 8621 Georgia Ave. 20910 Silver Spring, Md. The Motion of Ballistic Missiles
A
[It ,C r4lI I

Unclassified

NA

IVE! NOT F 5

7'p

of ripurt oand IfcitcI ivJ e datit,)

LI T "0 r4 1S

( Fi't

rIFAIo',

"Irddfle

fllNI

1d . IMN.- 1II*11"l')

L. S. Glover and J. C. 6
Ha.
EPrORT 0DTE

IHagen
7I. TOTAL NO. OF PAGES Tb, N0 O IEPS R

July 1971
CONT4ACT COR GRANT NO.

390
90, ORIGINATOR'S 4EPIORT N MIBER(SI)

20

N00017-62-C-0604 P o EC TN0.

G 1164
Wb. 0 TH ER HEPORT N OIB5 (Aniy a(her thin report)

'urnbe ti Ihetf rniy be asslaripi(t

d,
II DISTIRIBUTION STATEMENT

Approved for public release; distribution unlimited.


11. 5UPPLEII4ENTAIRY NOTES 12. SPONSORING MILITARY ACTIVITY

Strategic Systems Project Office


il. ABST14ACT

This handbook on the motion of ballistic missiles has been prepared primarily for use in training new personnel charged with the responsibility of evaluating the performance of reentry bodies. Aerodynamic fundamentals and characteristics of the atmosphere are reviewed briefly, followed by discussions of the effucts on ballistic missile motion of gravity, earth rotation, the atmosphere, and initial body rates. The anomalous motion resulting from various types of mass and/or aerodynamic asymmetries is discussed in detail. Simple solutions are given for the impact dispersion resulting from asymmetries and nonstandard meteorological characteristics at the impact location. Various trajectory simulations and flight test analyses are also discussed. Numerical examples illustrate the material.

-O-FF

UNCLASSIFIED
Security Claunification
14.
KEY WORDS

Ballistic missile Ballistic flight test data


Trajectory simulation Aerodynamic characteristics Training manual

Impact dispersion
Missile trajectory Numerical solution

:1

tt

*1
'1

I,

Security Classification

TG 1164 JULY 1971

Technical Memorandum

THE MOTION OF BALLISTIC MISSILES


by L. S. GLOVER and J. C. HAGAN

THE JOHNS HOPKINS UNIVERSITY *APPLIED PHYSICS LABORATORY 8621 Georgia Avenue, Silver Spring, Maryland 20910
Operaitnq under Contract WoCol 7- (3if.C..l6o4 with the Depdrinitirt of the. NA"'

Approved for public reless; dittribution unltimited.

......

ThEl JOHNS

HOP~i(NS UNIVENIiVI

APPLIED PHYSICS LABORATORY


IlLVR $PRIM* MAVI.6AND

ABSTRACT

This handbook on the motion of ballistic missiles has been prepared primarily for upe in training new per-

sonnel charged with the responsibility of evaluating the


performance of reentry bodies. Aerodynamic fundamentals and characteristics of the atmosphere are reviewed briefly, followed by discusEions of the effects on ballistic missile motion of gravity, earth rotation, the atmosphere, and initial body rates. The anomalous motion resulting from various types of mass and/or aerodynamic asymmetries is discussed in detail. Simple solutions are given for the impact dispersion resulting from asnmetries and nonstandard meteorological characteristics at the impact location. Various trajectory simulations and flight test analyses are also discussed. Numerical examples illustrate the material.

Lii

JOWN1 IhN

HOPKI4

UNIolv"ItV

APPSLIE

PHYSICS LABORATORY

ABSTRA CT

missiles This handbook on the motion of ballistic new pertraining in has been prepared primarily for use of evaluating the sonnel charged with the responsibility fundamenperformance of reentry bodies. Aerodynamic reviewed are tals and characteristics of the atmosphrre on ballistic effects the of discussions briefly, followed by rotation, the atmosphere, missile motion of gravity, earth

inotion resulting and initial body rates, The anomalous aerodynamnic asymfrom various types of mass and/or solutions are is discussed in detail. Simple

metries resulting from asymmegiven for the impact dispersion characteristics at tries and nonstandard meteorological simulations and the impact location. Various trajectory Numerical exflight test analyses are also discussed. amples illustrate the material.

,Ny

(:1
-

iii-

I
)*

APPLIED PHYSICS LABORATORY


SILVER S04114I MAIyLANO

THE JOHNNE HOPKINS

UNIVERSITY

CONTENTS

List of Illustrations List of Tables 1 2 3 4 4.1 4. 2 4.3 4.4 4.5 4.6 5 6 7 8 Introduction Aerodynamics Atmosphere Equations of Motion Effects of Gravity Effects of Earth Rotation Effects of Atmosphere
. . .
. .

. .

vii
xi 1.

5
. . . . . .

43 71 75 95 107 129 157 235 269 323 333 365 369

Effects of Body Dynamics Effects of Asymmetries Examples Dispersion


.
a

. . .

, ,

Trajectory Simulation . Flight Dynamics Data Analysis Constants and Conversion Equations Index ,
. .

.
.
.

1 l9.

Preceding page blank

tWIql JONhlO

'40lPM1PW6i UNIVIUSIITY

APPLIED PH',SICS LABORATORY

ILLUSTIIA TIONS i. 2-1 2-2 2-3 2-4 2-5 2-6 2-7 3-1 3-2 3-3 4-1 4-2 4-3 4-4 4-5 4-6 4-7 4-8 Variation of Viscosity with Temperaturc Variation of Viscwqity with .36 Temperature Cone Axial Force Coefficient, Flow, Cold Wall, 8 = 100 Cone Axial Force Coefficient, Flow, liot Wall, 6 = 10* Laminar 37 Turbulent 38 39 40 41 65 66 67 . . 89 90 91 92 .. 93 94 125 12(6

35

Pitch Damping Coefficient for 10 Semiangle Cone Roll Damping Coefficient, Cold Wall, 6 = 10* Laminar Flow,

Roll Damping Coefficient, Turbulent Flow, Hot Wall, 6 = 10 Typical Atmospheric Temperature . Profile and Altitude Regions Standard Atmosphere Characteristics Effect of Water Vapor on Density Range versus y 0 and V Initial y versus V and Range R

Flight Time versus Yo,

3000 nmi Minimum

Flight Time versus Range, Energy Trajectory Range Sensitivity to V Range Sensitivity to Y 0 Variation of II(K) with K Variation of I (K) with K
.

I'

vii

Preceding page blank

N,

IT4g jaid

MOR~NIN uNher.3TY

APPLIED PHYSICS LABORATORY


LVIIN 11PRIN41MARYLAND

IL.,L.f RATIONS (cont'd) 4-9


4-10 4-11

Variation of I, (K) with K


g g

1 27
.

versus K

128

Body Axis System

.155

4-12

Variation of J 0(W with X

155

4-13
4-14 4-15 4-16 4-17 4-18 4-19 4-20 4-21 4-22 4-23 4-24 4-25 4-26 4-27 5-1 5-2 5-3

Variation of A and A0 with X


Variation of X with D cot 0 Variation of G with X and D 0. 2

227
228 229 230 231 2312 233 2600 261 262 263 264K

Variation of F(X, D) with X, ID 0. 2 Roll Lock-In Criteria, Body A Roll Lock-In Criteria, Body B Roll Lock-In Criteria, Body C & Decay Characteristics Significant Events Required for g L pr and w
.

.259 . .

20'.20

versus Altitude

Conditions for Roll Resonance at h = 1) 000Feet Required for Lock-In and Spin tdrough Zero Roll Rate .. Boundaries of~ and for Lock-In at First Riesonance . . Variation of r with Initial Roll Rate mm. Limiting Values of KS. L. Limiting Values of A/ E Variations of E with h, Standard Atmosphere 1962
. . .

265
266

309 310 311

-viii-

SHC JOMNS HOPKINI UNIV9N*ITY

APPLIED PHYSICS LABORATORY


L

~ILv6A

IPUIWO

MAPYLSK

ILLUSTRATIONS (cont'd) 5-4 Dispersion Resulting from Wind and Deviation in Density 5-5 5-6 5-7 5-8 5-9 5-10 7-1 7-2 7-3 7-4 7-5 7-6 7-7 7-8 7-9 7-10 7-11 7-1 2 Dispersion Rcsulting from Meteorological Characteristics at High Altitude Probability of Encountering Zero Roll Rate Maximum Altitude for Zero Roll Rate Maximum Dispersion Resulting from Se and r/d Dispersion versus 0 0 Dispersion Probability . , 312 31 3 314 315 316 317 318

Bit Presentation Scheme, DITAP Flight Records . .353 Timing Traces, DITAP Flight Records Bit Presentation Scheme, DITAR Flight Records Timing Traces, DITAR Flight Records . Sample Roll Rate History Sample Lateral Acceleration History Sample Longitudinal Acceleration .. History Sample Pitch Rate History Angle of Attack Work Sheet . . . 354 355 356 357 358 359 360 361 362 . 363 364

Aerodynamic Pitch Frequency (Weathercock Frequency) Work Sheet * Amplification Factor Work Sheet Rotation of Maneuver Plane Work Sheet . . .
-

ix

I
* 'M[ JOMNIl UNIV[IRIYI't H'OPKIN3N SlPSlf,N

APPLIED PHYSICS LABORATORY


SILVIS MAi~llr,.A'i

TAB1,ES

2-1 3-1 3-2 4-1 5-1 5-2 5-3 5-4

Mass and Aerodynamic Characteristics 1062 Standard Atmosphere Constants for the 1962 Standard Atmosphere Reentry Characteristics Values of 13 Values of F

42 68 69 267 319 320 321 11322

Values of 15 Values of F

iK

ix

Sxi

Preceding paga blank

........................ ........................................ r:.

1.-3

APPLIED PHYSICS LABORATORY


111111021 111IMS MARYL.AND

~;

1. INTRODUCTION

APPLIED

-HYSICS LABOrATORY

The term ballistic missile as used herein describes an!j body in motion that is unpowered and uncontrolled. The motioa may occur in either the atmosphere or the exnatmosphere. However, most of the interesting motion phenomena occur wiithin the atmosohere, and most of the discussions concern reentry-type vehicles entering the atmosphere at speeds from 10 000 to 25 000 ft/sec.
'l l( p'iimar 'yv purpose_) oft thiL, rclp()rt is to pre sent

material suitable for' training new personnel charged with the responsibility of evaluating the performance of reentry bodies; therefore, the subject matter is treated as simply as possible while retaining enough detail to explain the missile behavior. Although the material is written primarily for personnel unfamiliar with the detai's of missile motion, it provides experienced personnel with readily available equations suitable for estimating order-of-magnitude effects of phenomena frequently encountered in flight test analysis, performance, and preliminary design work. A brief discussion ot the aerodynamic characteristics of missiles and the nature of the atmosphere is followed by a discussion of vehicle motion as affected by gravity, earth rotatior., the atmosphere, and initial body rates. Also included is a detailed discussion of the anomalois motion resulting from various types of mass and aerodynamic asymmetries. Simple solutions are given for the impact dispersion resulting from asymmetries and nonstandard meteorological characteristics at the impact location. Simulations required for various types of studies and flight test analysis complete the subjects discussed. Numerical examples illustrate the material presented. In some sections (particularly parts of Section 4) the material is new and is given in considerable detail, Since some readers may not be interested in details, a summary of the material covered and the major conclusions derived is given at the beginning of each section (each subsection for Section 4).

Preceding page blank

"~

~~~~~

--..

.-

TMa JOHNS NMOPKINO UNIV90SITV

APPLIED PHYSICS LABORATORY


BILVIS OSPONO MARYLAND

An attempt was made to use symbols and units of measure currently used by specialists in the various fields covered in this report. As a result, a particular symbol mayhave different definitions in various sections. A complete list of symbols is given at the beginning of each section (each subsection for Section 4), and conversion factors for units of measure are given in Section 8.

-4-

___________________________________________________________I

~APPLIED PH4YSICS

flu .AewD nrs "ANSurNvmweln


LAISCRATOftY

2. AEAODYNAMICS

*jI Ii

THE JOHNS NiOPKINi UNIVE"ISIY t

APPLIED PHYSICS LABORATORY


lllILVE.l SPR,NG MAmv.ANO

SYMI3OLS FOR SECTION 2

Symbol A CA CAB CA F CA

I)efinition axial ro''rce axial force coefficient base drag coefficient friction drag coefficient pressure drag coefficient

Typical. Units pounds

P
c, g. C p C Cm
0

refers to center of gravity rolling moment damping coefficient pitching moment coefficient pitching moment coefficient at 0-

Cm q C CN CN el

pitching moment damping coefficient dC m /d&ata"= 0

normal force coefficient 00 dC N/dWat refers to center of pressure total drag drag resulting from skin friction reference diameter Knudsen number (see Eq. (2-7)) lift or body length pounds feet pounds pounds feet

c. p. D DF d K L

Preceding page blank

Jt

THt JOHNS NOPING UINIV"mUIT

APPLIED PHYSICS LABORATORY


MILVIR SPRING MANVLANO

Symbol
L(p) I M

Definition
rolling moment resulting from roll rate (roll damping moment) moment arm for asymmetric force Mach number = V/Ior

Typical Units lb-ft


feet

pitching moment M(q) N N N


s

lb-ft lb-ft pounds pounds pounds lb/ft2 lb/ft 2 lb/ft 2 rad/sec rad/sec
2

pitching moment resulting from pitch rate (pitch damping moment) normal force

asy

normal force resulting from an asymmetry normal force for a sym.metrical vehicle ambient pressure cone surface pressure flap surface pressure roll rate pitch rate dynamic pressure, (y/2)PM (1 /2)pV2 gas constant = 1716 Reynolds number, oVL/14 = VL/h reference area static margin surface area excluding the base area absolute ambient air temperature reference temperature for viscosity vehicle velocity
01l

P P PF p q q 1R Re S S.M. w T TO V

lb /ft2 ft /sec ft 2 ft 2 OR OR
ft/sec
-i

-OR

"-8-

A
a.4)

TYIO JOMW NO*tINS

UIV||glvTl

APPLIED PHIYSICS LABORATORY


SILVIft BP*N;% MANU6AN6

Symbol X

Definition distance from nose to a point on the body centerline

Typical Units

feet feet feet

x x

g.

distance from vehicle nose to the c. g. distance from vehicle nose to the c. p.

AX
Y (w t

c. p. c. g. distance normal to the body surface vehicle total angle of attack local angle of attack of specific heats
-

-x

feet
feet radians radians

v yratio
8
(

1. 4 radians radians slug/ft 3 slug/ft-sec slug/ft-sec ft 2/sec lb/ft2 degrees

cone half-angle asymmetry deflection angle ambient air density

M Ao

dynamic viscosity dynamic viscosity at T kinematic viscosity shear stress

AO

body bend angle

"-9-

TMK Jaws MOPIINGI UNIVIRSITa*1

APPLIED PHYSICS

LABIORATORY

SUMMARY AND CONCLUSIONS

The aerodynamic forces of interest for reentry body studies are discussed, The significance of the aerodynamic coefficients is reviewed, and equations are presented for estimating the magnitude of the coefficients for a sharp-nosed conical hody. Several types of asymmetric cones arc nlso con Hidn(rcd. Numerical examples are given for a 10 serniangle cone having a length of 10

feet.
The major conclusions are: 1. The following aerodynamic forces and moments are important in the analysis of motion of a typical reentry body: v Force perpendicular to tho body centerline (normal force) and resulting moment about the body center of gravity (pitching moment) Force parallel to the body centerline (axial force) Moments resulting by virtue of body pitch, yaw, or roll rotational rate (pitch, yaw, or roll damping moments) Forces and/oi moments resulting from body asymmetries.

* *

2. For uncontrolled vehicles, it is important that the normal force act through a point (center of pressure) aft of the center of gravity. In this case the vehicle is said to be statically stable. The ratio of the distance between the center of gravity and the center of pressure to some characteristic body dimension (for example, length) is called the stability nialgin and Is usually expressed as a percentage of the characteristic length. 3. The aerodynamic forces and moments are represented by dimensionless coefficients, thereby facilitating evaluation of the forces and moments over a wide variation in environmental parameters (velocity, altitude, etc. ),

Preceding page blank

Twx

JOHNS MOPKINI UNgVDII4Y giILvIIm SPlllii6.


M&WVkbIIII

APPLIED PHYSICS LABORATORY

4, In general, the coefficients are functions of vehicle angle of attack and the similarity factors, Mach number - an indication of the effect of compressibility of the air, Reynolds number - an indication of the nature of the boundary layer on the vehicle, and the Knudsen number - an indication of the size of the vehicle relative to the mean free path of the ambient air particles.

-12-

1
":..' ~~~~~~~~~~~~~~~~~......................'.

......

... '..,..', 4'i :l"...!z.t.',!! .. ~.v- -'..I

:"

?MI' JOM41S HOP1q.N

UNIVR~P91?

APPLIED PHYSICS LABORATORY


Si VI *PRllhO MAHYLANt,

2.1 Aerodynamic Forces A body moving through air experiences forces aerodynamic forces - resulting from the motion. The aerodynamic forces of interest for a reentry vehicle are: 1.
2.
A.

Axinl force, Norinmal force,


'

itch damping force, toll damping force.

The axial force (for a symmetrical vehicle) acts along the body centerline as shown in the sketch for a cone.

C.

".

This force is produced primarily by relatively high pressure acting on the conical surface (pressure drag) and In addition, a force produced by air viscosity (skin friction drag) exists on the conical surface.
A normal force (normal to the body centerline)'"

relatively low pressure acting on the cone base (base drag).

exists when the body centerline is not directed along the

II

-13-

YHI JOHNS HOPKIN8

,NIVINIITY

APPLIED PHYSICS LABORA'rORY

velocity vector, The angle between the velocity vector and the body centerline is called the angle of attack, . The normal force results from high pressure acting on the windward side of the body compared to low pressure acting on the leeward side. The integration of this pressure difference over the cone surface is the normal force. The pressure differential across the body alsc produces a mornent about the body center of gravity, Poor convenienctin analysis, this aerodynamic moment is computed as the product of the normal force and a moment arm (X is the longitudinal location at 1h6ch X1 . ), where X the fotal normal foice acts and is called the center of pressure and Xc. g. is the longitudinal location of the center' of gravity. Assume that a vehicle has its center of pressure forward of the center of gravity and that some disturbance causes an initial angle of attack. This angle of attack results in a normal force, Since the normal force is forward of the center of gravity, the aerodynamic moment produced is in a direction to increase the angle of attack, This process continues, and the missile eventually tumbles. Similar reasoning will show that, for a center of pressure located aft of the center of gravity, the aerodynamic moment will always tend to cancel a perturbation in &, Therefore, a vehicle having a center of pressure forward of the center of gravity is said to be statically unstable; a vehicle having a center of pressure aft of the center of gravity is said to be statically stable. The value of the ratio(Xc, p. - Xc. g,/L w AX/L is called the static margin. The margin must be positive for uncontrolled vehicles and for high performance vehicles is usually 3% to 5%. The use of the word "normal" to define the force In the plane of ;T and perpendicular to the body centerline is used here to be consistent with the usual terminology found in the literature. It is hoped that there will be no confusion with telemetry terminology where "normal" force and 'lateral" force are the two components of the In an attempt to avoid normal force as defined above. confusion, in some sections of this report we have used

-14

TNS JO NOHOPKINS.rNIIvuuTY

APLIED PHYUICS LABORATORY


BILVIN OP01.14 &iAP.I6ftD

the term 'transverse' force to mnean the force in the plane


of a and perpendicular to the body centerline. The pitch and roll damping forces exist by virtue of the body pitch and roll rates, respectively, and therefore are called dynamnic forces. Consider a body pitching about Its center of gra,-vity with rotational velocity, q, The pitching motion produces a locail velocity at X equal to 1X- X )q. The distribution of velo.zity along the body is shon n the sketch,

Mg.
The local velocity distribution produces a local anrgle of attack, &I, whose magnitude is given by the equation:
(X
-

1
*

X)q

The local angle of attack results in a. local normal force which, at all points along the body, tends to resist the pitching motion. The Integral of the local forces (the damingforce) is tosmall to be of significance in ballisoet h apn tic missile aaye.Hwvr which is the integral of the local forces multiplied by their moment arms, is very important in some portions of the trajectory as will be shown later.

15 -

INIiV2 1ItY MOPKIN&I THC JO~HN% APPLIED PHYSIC5 LABOnATORY MARI&NIA t$11V SIPRING

Tile example cited above i.s just one type of damping effect. Any force that exists by virtue of the pitch rate is a pitch damping force, and these forces do not always re.sist body pitch motion. The forces that do resist the pitch motion are said to be stable forces (or moments), and those that reinforee the pitch motion are said to be unstable forces (or moments). The only roll damping force on a symmetrical body of revolution is the force produced by the viscosity of the air, (skin friction). This force always tends to retard the roll rate. Ideally, it would be desirable to calculate the oerodynamic forces accurately by theoretical means. However, thio is nut always possible, and it is necessary to measure the forces in ground tests by blowing air past a fixed body (wind tunnel), or by propelling the body through a stationary instrumented region (ballistic range), or sometimes by using a combination of both techniques. Bconomically it is not possLble to obtain measurements on the actual vehicle for actual reentry conditions of ambient pressure, temperature, velocity, etc. To circumvent this difficulty, the aerodynamicist defines the forces nnd moments in terms of aerodynamic coefficie'ts which npply over a wide range of environmental conditions. The following coefficients are of interest in the study of renritry body motion:

A - CA S
N m CNqS M(q) _C (-. iSd

(2-2)
(2-3)

(2-4)

V 1 p
-

16

?b41

JOH$1404SHOkpIN UN'VtN6ITV

APPLI[0 PHYSICS LABORATORY

where q at(1 /2)pV2 (,Yf2)pM 2 an(_ is called dynamic pressure. Tile bar over the q is deleted In somic sections of this report where there is no possible confusion between the symbols for pitch rate and dynaminic pressure. The reference diniensions S anrd d niay be iany convenient dimensions. For a cone, they nre usuanlly tnken as the ba sec ja0, Rca d h118 edin ( met. er, and thusu~ referen~ce dimen-

rrom dimensional nnalys Is, it may he shown that for a given ;; thle "'rody"-namic coe-fficients CA, CN (and

n geometrically similar model (externnal, shape) provided


certain modeal stimilarity f'actors are tile same as those

'

for the actual vehicle.

The nimajor similarity factors ofI

interest for, the reentry body are:

TieIch numbor.

is tile ratio of the vehtcle

dspeeds

velocity (V) to the velocity of a sound wave (V-ylT). "'here are four significant Which number re,,gimes. At subsonic (M ft0 to 0. I), the acrdynnai coefficients are weakly dependent upon M. Tho transonic Mach number regime extends from thle Ma~rch number at which a local Mach num-ber over the body fi1rs~t becomes sonic (MI 1) to the Mach nui-nbr, nt which.1 the flowfield over the hody is predominantly supersonic. This Mach number ra-nge speed results in the, appetirance of shock waves, Trhe strength and location of the wnveF are highly sensitive to changes in Mach number. Therefore, in the transonic speed range, thle aeurodynnmic coefficients are usually sensitive to change in M. In the supernonic speed range (M. P 1. 5 to (6), thle shock wave stteength Is dependent upon Mach number but the shock location is somewhat insensitive to changes In Mach number so that the aerodynamic coefficients are moderately dependent upon M. The effect of

-17-

114l JOamNm OPRIN9 UNIVSII6TW

APPLIED PHYSICS LABORATORY


*IhVOR @POllS 14A.y lkO

speed aerodynamic heating 12 moderate in the supersonic aerodynamic the 6), > (M range. At hypersonic speeds and coefficients tend to be relatively insensitive to M aerodynamic heating is severe. The Reynolds number, pVL/M = VL/v, is indicaof the air, tive of the importance of the dynamic viscosity viscous of result a As M4 (or kinernntic viscosity t). in the vicinity forces, a region of low velocity is produced below) (discussed flow of the body. In fact, for continuum at the zero is vehicle the to the velocity of the air relative called is velocity degraded vehicle surface. The region of the to adjacent air of layer the boundary layer. Since the an or r, stress, shear a body "sticks" to the surface, element of friction drag, dD, results, tending to reduce the velocity of the vehicle,

-I

Y (NORMAL TO S'URACE) --------j

LAMINAR

-TURBULENT LAMINAR BOUNDARY LAYER THICK(NESS TURBULENT BOUNDARY LAYER ,' BODY SURFACE "THICKNINI

The total friction drag is given by: DF = .II" dSw

~,dS dr
TA (" dy

(2-6).,

*cosity
,

Therefore, the viscous drag is proportional to the dynamic wall. The visviscosity and the velocity gradient at the only temperature air of function a nearly is very
-

''

18

}i

?WEl JO~kuU MO,UOD~til. uMIIKftIJ

APPLIED PHYSIC

LAMOHATORY

and is given by Sutherland's equation plotted in Figs. 2-1 and 2-2. The velocity gradient is a strong function of the Reynolds number. At very high Reynolds numbers, the flow in the boundary layer contains much eddy or vortical motion. As a result of mixing, the high energy air near the outer porLion of the flow is transported in toward the surface. Since the layer of air adjacent to the wall sticks to the surface, this type of boundary layer has a very large velocity gradient at the surface and the skin friction is relatively high. This type of boundary layer is called a turbulent boundary layer. At sufficiently low Reynolds numbers, the eddy motion within the boundary layer disappeari, and the velocity gradient at the wall and skin friction become much less than those for Lhu turbulent case. This type of boundary layer is called a laminar bounda,'y layer. At low Reynolds numbers the flow tends to be laminar; at high Reynolds numbers the flow tends to be turbulent. The Reynolds number at whinh transition from laminar to turbulent flow occurs varies with body shape and type of trajectory. A typical value of Re at which transition begin- to occur on a slender conical body

is 107.
The surface heating rate is proportional to the skin frictioi and therefore is greater for turbulent flow than for laminar flow. In fact, the heating rate in turbulent flow may be nearly an order of magnitude greater than that for laminar flow. The Knudsen number is the ratio of the mean free path of the ambient air particles to a characteristic dimension of the reentry vehicle. This ratio may be expressed in terms of Re and M by the relation K = 1.49 R eM (2-7)

For the very low values of density that exist early in the reentry phase, the particles of air are spaced at distances that are large compared to, say, the body length

19 -

tHR JOHN$I NHOPINf UN1VCSIIty

APPLIED PHSICS LABORATORY


*ILVlI* *pegINO MAPYLANO

or diameter.

An air particle bouncing off the body is not

likely to encounter another air particle. This region of flow is called the free molecular flow regime. At a somewhat lowtr altitude, the air particles have frequent collisions with other air particles in the vicinity of the reentry body. However, there are insufficient particles in the film of air adjacent to the body surface to cause it to stick to the surface. Therefore, although a boundary layer exists, the film of air adjacent to the wall "slides" along the body surface, This flow regime is called the slip flow regime, At still higher densities, the air particles at the surface stick to the surface, is called the continuum flow regime, This flow regime

The characteristics of the aerodynamic forces and surface heating depend upon the type of flow that exiots on the body. There are no sharp boundaries between the various types of flow, but approximate boundaries are as follows: Continuum to slip -1i0
2

Slip to transition

W 11 -1

m
Transition to free molecular M R 10 e

10

The transition regime considered here is a transition in basic flow regimes which occurs at Knudsen numbers of approximately unity. This transition should not be confused with transition from laminar to turbulent flow in the boundary layer. The Mach number, boundary layer, and basic fluid flow regimes are shown for a typical reentry body in Fig. 4-21, page 260. Note that practically all the reentry ,

20

I * '

i I~-I i.i,:lt *l r A

THI .. OWNS IIOPKIN, U vI v APPLIED PHYSICS LABORATORY

phenomena for which aerodynamic data are required occur in the continuum flow regime; only this regime will be considered in the remairnder of this report. Sometimes the lift nnd drrnf4 components of the for nnnlysis. The relntionship between the liift-dmtn g systclin of forces nnd the normal force-axinl force system is shown in the sketch.

nerody.iamic for'ce neiroqutired

SN
A
a4

From this force diagram, the lift and drag coefficients may be obtained from the following equations: C

L,

cos

- C

sin a

(2-8) (2-y)

CD =CA cos e + CN sin

Differentiating Eq. (2-8) with respect to c, and noting that the variation of CA with ey is small, the slope of the lift curve at a = 0 is given by the equation: CL =CN -CA (2-10)

-21-

__

"-"%,. .. t'.

tHil JOHNS HOpKININ jNIV IIv

AM-t.ILD PHYSICS LABORFATORY


GI-10N SPIIN. M.S-tAN.,

2. 2 G(enural Chnaincteristics It is beyond the scope of this r'eport to provide proce(duI'cr5lo'f estilmting the naerodynalmlic coefficuients for n body of arbhira'ry shape, Ilowever, a few re mareks on the

genernl behavioi' of tie coefficients nae given in this seetio)n, and :ppi' xini mte equa tions applicanle to n shn rp nosed CWne :i 1,fiVe.ii ill SCct ea 2. 3. Axial 'rcee. For most bodies of interest, the axial foorce coefficient is weakly dependent upon c for small values of ' v(10 to 2*), strongly dependent upon M, andt moderately dependent upon R.0 . For a sharp slender cone, the order of magnitude of CA is 0- 1. Normal Porce. The normal force coefficient is strongly dependent upon rv and M and nearly independent of R1. For small values of fy, CN varies linearly with n so that CN = CNrY. For a sharp slender cone, CN a. 2 per radinn. Center of Pressure. In general, the center of pressure is strongly dependent upon o' and M and weakly dependent upon Re. However, for small vwlues of rY, Xcp P is independent of ry. For a sharp slender cone,
Xe p:, (2/3)L,

Pitch lDnn-ipng. The pitch damping coefficient is strongly dependent upon M, weakly dependent upon & for small values of ry, and nearly independent of Re. Fora a sharp slender cone, a typical value of Cmq is -1 per radian. The negative value of CnMq indicates a stable damping coefficient. 1oll Damping, The roll damping coeffici.nt for a symmetrical vehicle is very small, negative, hltiily dependent upon Re and M, and weakly dependent upon r for small values of . For I a cone, a typical vwlue of the roll clamping coefficient is -0. 003 per radian,

:1

22

I
tk " "'4

'L

r\
?HU[ JO94NS NOP4dINg) UHIVEMITYI

APPLIgED PHYSICS LABORATORY

2. 3 Approximate Equations for a Sharp Cone For a major portion of the reentry trajectory, the Mach number is in the hypersonLc regime. The aerodynamic coefficients are nearly Independent of Mach number, and good approximations to the aerodynamic coefficients for sharp nosed cones are given by relatively simple equations. The skin friction drag coefficient, CAF, is very sensitive to the cone surface temperature, to the type of boundary layer, and to surface roughness. To provide some idea of the order of magnitude of the coefficient, CAF, without becoming involved in details, equations are given below which apply to laminar flow with a cold wall (typical of early reentry conditions) and to a turbulent flow with a hot wall (typical of late reentry conditions), A smooth wall was assumed. An estimate of C, is given in which it is assumed irected along the vector Is that the shear stress + pr where pr is the circumferential velocity of a point on the cone surface. The circumferential component of Ir is proportional to C1 The following equations are applicable to a cone of half angle 8:

CN

- 2 cos N2

28

(2-11)

Xc.p.

2sec

8 L

(2-12)

=C

AP

+C 2

AB

+C

(2-13) (213
(2-14)

2 sin
1,

CAB

M.

(2-15)

-23-

UNIVINSITY ?UK JOHNS HOPKINS APPLIED PHYSICS LABORATORY

CA

1. 9

tai 8

2+sin 2)Cosl

(2-16)

Laminar flow, cold wall

CA2Co 8 tan

0.___ (1 +

M _-;n 6

cc)1.
r

CA

01

Bodies. 2 2. 4.ymeri to b+ syterin Boie inee

alay

(2r1

etri aBodieos Considym

hc teaymtrcfrei

sroduce

lcate proubrnces ths small byevr

diastru neavte

end of the body, as shown in the sketch.

-24-

TIMiR JOHMN NO$ H K MO NIS|V9IT?

APPLIED PHYSICS LABORATORY


SILVIO Se SIN4 MAPVLANO

At a 0, the protuberance results in a downward asymmetric normal force, Nasy. This force causes the nose to pitch up, producing an angle of attack. The angle of attack produces a normal force, Ns. The vehicle continues to pitch up until the moment about the center of gravity is zero. The zero moment condition is called the trim condition, and the values N and re that exist at this condition are called the trim normal force and trim angle of attack, respectively, It is convenient to define another dimensionless aerodynamic coefficient called the pitching moment coefficient, Cm, such that the moment about the center of gravity is given by!

M=CM q Sd.

(2-20)

From the above sketch (using a nose-up moment as positive): asy 1 (s c.p. c.g.

The total normal force is given by: N Ns - Nasy (2-22)

Writing Eqs, (2-21) and (2-22) in coefficient form:

Ns i asy q Sd
Ns IS--(IRS Nasy S

N (Xc -X ) s c.p. c.g. d qS


,

(2-23)

2-24)

The term, Ns/lS, is the CN for a symmetrical vehicle. The Nasy term is usually negligible compared to Ns/qS

-25-

THl JOHN*41 hiJPKINi UNIVRIENITY

APPILIED PHYSIC,, LABORATORY


216YINSPmIN4 MARh.At.O

and may be ignored in Eq. (2-24). However, the effect of this term on Cm cannot be ignored. Let

asylaC
qSd Then (X x

C
0

C
N

C. .
d

(2-25)
then Eq,

If the body has a linear variation of CN with , (2-25) may be written:

" C

p. d

m 0
If Cmo and X., p, are independent of 9, linearly with a.

(2-26)

then Cm varies

Within the limitations imposed by the assumptions listed above, typical Cm characteristics for a stable symmetrical and a stable asymmetrical vehicle are shown in the sketch, page 27, Note that C is the value of Cm at o 0, and TR is

aTR is the value of ey at Cm = 0. obtained from Eq. (2-26) as:


Cm

The value of
C

m 0 -x

i 0 Z2-7 (2-27)

~TR

(X

CN

-26-

TH

JIOHISi "O

INs UNIVOWAIIY h.A

APPLIED PHYPICS LABORATORY

S*I4.VlUR UkHU UAUI

CL

ASYMMETRICAL VEHICLE

, -.

SYMMETRICAL VEHICLE

I'

In terms of the static margin defined previously: Cm 0 -R

TR

(L/d)CN S,'M.

o(2-28)

In some reports the term Cn, is used. with respect to . Cm - CN Therefore: I

An equa-

tion for Crma may be obtained by differentiating Eq. (2-26)

(2-29) con-

To provide a feel for the magnitude of Cm.

sider the flap-type protuberance discussed at the beginning of this section.

27

iN

Thi

JOINI 4OPKINS UNIVURSIIV &ILvIR SPRINGi MARflV6AND

APPLIED PHYSICS LANORATORY

'I

a 5 P1

ley
qA

'

-~

.1.

The pressure acting on the flop and cone surfaces may be estimated using Newtonian theory which, in the simplest form, states that (for zero angle of attack):
PF "2
q

A"1

2 sin 2 ( + c)

(2-30)

=2

sin 28

(2-31)

may be divided The force resulting from the asymmetry into normal and axial force components. For small values of 6 and (, the axial component Is small compared to the normal component. Neglecting the axial component, the total increment in normal force is:
Nasy-SF(PF

(2-32) i

Then: N M qSd qSd


I S *IP

IS d

28

3) (2q

- 238

- p.

'

Tr" JOHN$ MOPKINN UNIV910ITY

APPL.OID PHYSICS LAgORATORY


11 6%.Vt4t ING M&AYI.A4ND

Using Eqs. (2-30) and (2-31), and assuming 6 and ( to be small:

Cm
0

-100, 1/d 1, ,SF/S

(2-34) (2T 0.1,

As an example, let8 and I r Cmo

20. For this asymmetry, Eq. (2-34) shows that 0.0027. Another type of asymmetry for which Cmo can be

readily evalUL:Lcd Is one in which the body centerline is the arc of a circ*le, This type of asymmetry, for example, is an approximation to a distortion resulting from lateral loads or heating effects. For this asymmetry the moment is noarly a pure couple and Cmo may be expressed in turms of the change in slope of the body centerline from nose to base.

007

At high hypersonic Mach numbers:

mo

O 0.008 Le

(2-35)

where A is in degrees.

-29

Ii

..

. ,. .... .... .. :.,..

.H

fTl

JIOHNSIHOPKIPN UNIVERIIIv
SILVIN SPMI2N MANYt.AND

APPLIED PHYSICS LABORATORY

2. 15 Examples
A typical reentry body will be used in many examples In later sections of this report. To illustrate the material in the preceding subsections, the aerodynamic data required for subsequent examples will be computed. Th r typical body is a 10-foot-long cone with a base diameter of 3. 5 feet. These dimensions correspond to a cone semiangle of about 100. The mass characteristics are assumed to be as follows: Weight X 1000 pounds
= 6. 27 feet aft of the nose

Pitch moment of inertia about the 2 center of gravity t: 300 slug-ft


2 Roll moment of inertia - 30 slug-ft

We will use a reference length ence area S of 9. 65 ft 2 .

d of 3. 5 feet and a refer-

At hypersonic speeds and for angles of attack close to zero, the aerodynamic coefficients are as follows: SC.
, Xc p.

2 3 se c 1oo0

10 -6,87

feet

(2-12)
(2-11)

N-

2 cos

1.94 per radian

It is observed from Eqs. (2-13) to (2-17) that CA is a function of M, Re, wall temperature, and type of boundary layer. The valves of CA for two combinations of wall temperature and boundary layer type are shown in Figs. 2-3 and 2-4. The laminar flow, cold wall case is typical of conditions that would exist early in the reentry flight, say above 100 000 feet; the turbulent flow, hot wall case is typical of conditions which would exist loter in

.I

I
THE JOHN@ HOPKINS UNIVR"4IY

APPLIED PHYSICS LABORATORY


SIIA 1 SPRImNSl MANYIANS

flight, say below 100 000 feet. As will be shown later, most of the phenomena of Interest occur for values of He from about 106 to 108. For these values of it

CA ft 0.i 1

A
At M = 10 approximately (60% Ot' CA results from presrure drag (CAP), 30% from skin friction (CAF), and about 10% from base drag (CAn). The percentages vary somewhat with Re, M, and type of boundary layer, but the given values are typical of the relative importance of the three sources of axial force for low-drag reentry bodies. For blunted bodies, the pressure drag may account for nearly 100% of the total drag. Values of Cmq (Eq. (2-18)) are shown as a function of X , /L i'n Fig. 2-5, and values of C,(Eq. (2-.19)) are shown in 'Figs. 2-6 and 2-7. p In the examples included in subsequent sections we will use typical values of the aerodynamic coefficients. The values of the rmass and aerodynamic characteristics that will be used are shown in Table 2-1.

31
,i

-31-

.'aNIV~qiTY APLIEDl PYSICS LABORATORY DILtY1 SPRIh MAMILAND

Til,

JOHNN* "O"'B

aN

BIBLIOGRA PHIY

2-1

S. A.

Schaaf :tlid L. Talbot, tl-ndbook of SuperMechanics of Rarefied Gnses, NAVORDSYSCOM Report 1488 (Vol. 5), 1"',n!o,

[onic Aei.'odynaIAC8, Sectiun 16i,

2-2

M. Tobak an, \ 1, . Wel'rend, Stability Derivatives of Cones nt Supersonic Speeds, NACA TN T776-8, Se.pte,.,L.,.t_' ,

Preceding page blank


-33I'

_"_'_

_-,_..,

=,

iI

A12P L IED PHYSICS LABORATORY

SILVI op-Tp

IA

LU

JI
( I-

u...
I0 2L

I II

rfu

*35

Preceding page blank

?"t JOHNO HOPKINOS ISkLYS *eSh~

N141VCSIIY

APPLIED PHYSICS LAtORATORY


U&*$.. ASN.A

'I
4
-

3,1 X103

SLUGFTSE

!iQ2

-To

619 o R

--

SUTHERLAND'S FORMULA
4+

0 (rT~ ~TM AND To IN R"-,

191

0I

oi

1Qf3200

12

l96.7/TO
198.71

400

R00

12DO

1(oo

2000

2400

2800

3600

4000

TEMPERATURE (CR)

FIg, 2.2

VARIATION OF VISCOSITY WITH TEMPERATURE

-36-

,I

I...
_ _

,i

YMI jhWId

NOl.w6

UWi'4aus' 'V

APPLIED PHYSICS LABORA-r)H"

0.42

0.3

0.2

1
k. A
106

0104

Sig, 2.3

GOLD WALL, 8 ICIENT, LAMINAR FLOW, CONE AXIAL FORCE COEFF

10.

-37-

TMK iJOmNs MOPMKINI UNIVIrS1IITY

APPLIED PHYSICS LABORATORY


SLVIO 5,0 mpaN MAR6A~ND

',2

-.......

.....-

I..... ,. .

......

r.....

r-i T- 'r ....... -

...

. ----i M" 10 20

... '7 *

< 0.1

_____n_____

10 5

10 7

10u

10t

Fig, 2.4

CONE AXIAL FORCE COEFFICIENT, TURBULENT FLOW, HOT WALL, 6 100

38

4
I...

- 38-

S,'.,________,_,_._,___.

TII

JOHNS MOPKINS UNIVUIIYTY .SLVIA 2P..IQ MARYLAND

APPLIED PHYSICS LABORATORY

*1

.1.2

-0.4

I 0.54

I 0.58 X cg./L

I 0.62

I 0.66 0.70

0.60

FIg, 2"5

PITCH DAMPING COEFFICIENT FOR 100 SEMIANGLE CONE

-39

S......
.

..

....

....

.. ",

',',s..,' , . -. ,','-;-:..t"..

'.

. op'AIToVY IJHN$

A'

o *La,

-_P

,,e',.A TO

.0,04

-0.04 o.030

M - 20

l .o.02 0

iOoo 0~ 10 0 9 F,2-6 ROLL DAMPING COr LAmiaR FLO, COLO NALL, 8 1I

40'

THE. JOHN$l HOMrKNI NIV9M01161TYJ APPLIED PHYSICS LABORATORY


SIhlYSS *pPIMIS MAU~hANS

0.0003

T'

"

-0.002

.0,00`1

lo'

10

FIg. 2-7

ROLL DAMPING COEFFICIENT, TURBULENT FLOW, HOT WALL, 6

100

-41-

i
S
" -' ., :.:. r ',

. .. ...
,, "r. , . h." "" ,-' ""'.g , .: r...l m .I,, 1 .

THi JOHNS HOPKINS UNIVliUMIT

APPLIED PHYSICS LAUONATOV


Oiwvlm NPimI MARVLANI,

Table 2-1
Mass and Aerodynamic Characterlistics Masi3 Characteristics L 10 1 feet ft 2 S 9, 65 d m 3. 5 feet X g, . ,27 feet 6 Weight (W) - 1000 pounds Mass (m) - 31. 1 slugs 2 30 slug-ft x (Iv) Inertia of RAI moment 300 slug-ft z Pitch moment of inertia (Iy I)
ly
a2

0. 99

Ix

.2

0.079

Aerodynamic Characteristics C 2 per radian C a 0. 035 per degree


l~c'P.
.

Combined Parameters l01b/ft 2 W = 100S u A A


S.M.

6. 67 feet -1 per radian -0. 003 per radian

=-

"

/4%

CA -0.104 Cm C, p AX

0q 14,

-42-

*;,.

APPLItD PHYSICS LABORATORY


SILVli OPPING MhiV6A-0

?Hl JOHNlS UOmIms UNlIVIryIY

I,

3. ATMOSPHERE
4-

AA

[
!: [ - 43 -

lHIr

JIONNSI
JO

'40pINSI

UNIVSNS,?YlT
"NIVIPSI.Y

APPLIED PHYSICS LABORATORY


SINK
HOPfINS

SYMBOLS FOR SECTION 3


T_. 'it/sec pounds pounds2 ft/sec feet feet feet ib'ft c nl nits

Symbol a

Dinlittion speed oL sound

Napierian base
c

Coriolis force horizontal force resulting from a pressutre gradient acceleration of gravity scale height, RT/g 0 geometric altitude geopotential altitude static pressure of dry or moist air partial pressure of dry air saturation pressure at T slb/ft partial pressure of water vapor gas constant for dry air radius of earth

g H h

h
P P p p R

CT

2 2

1b/ft lb/ft2 2 l/t ft2/sec feet ft /sec

~2

R RAOB's

ROCOB3's

gas constant for water vapor weather measurements in which a balloon is used as instrumentation carrier weather measurements in which a rocket is used as instrumentation carrier sea level temperature of dry or moist air dew point temperature

S. L. T T

R
OR

45-

Preceding page blank

TH~r JONNI 9IOpMINII UNIV|ftI1IT

APPLIED PHYSICS LADORATORY

Symbol t V
V W dx, dy, dh Z c

Definition time missile velocity


wind speed weight of nn elermentary volume of nir dimensions of an elementary volume of air refers to time at the zero meridian change in density defined by Eq. (3-22) ratio of specific heats relative humidity or latitude

Typical Untis seconds ftsec,


ft/,ccqC

pounds feet

percent

v
0

degrees feet ft2 /sec slug/ft 3 slug/ft 3 rad/sec

mean free path kinematic viscosity density of dry air x P/RT density of moist air earth rotation rate

P
pr

Subscripts In o refers to a reference altitude or to altitudes at which weather measurements are obtained refers to sea level conditions

4'

APl'. itD PHYSICS LABORATORY


3ILVUN
SPANQ~ MARYLA"D

SUMMAARY ANF) CONCLUJSIONS

The atnisphore from se,- level to urhout 400 000 feet Plays '1n imlpoi't:uit vole ill the moition it'~rumiir V vehicles. In this section the generna mi tur1 , of ti n till sphere Is considered, and equations required for studies concerning vehicle motion Lire discusised. The vnria.blvs of most concern are atmospheric, density, temperature, wind, and moisture profiles (vtitrat ions with altitude), The 19~62 Standard Ati-osphere i~s used in many anialyses. Some of the charactvristics of this atmosphere are prosented, and differences from- othow types of sttandalrd days are noted. Tho major conciusionis iru1 . Most phenomena of interest in the analysis of reentry body motion occur in an altitudo region cnfled the homosphere (son level to approxiimately 300 000 feet), In this region the atmosphere obeys two laws of considerable importance. The perfect gas law Provides a r'elationship between pressure, temperature, and density; the perfect gas law in conjunction with the hydrostatic equilibriumnlaw provides a relationship between density, temperature, and altitude, The hydrostatic law 1is frequently used as an interpolation equation for tabulnr values of density versus altitude, permitting relatively few volues of the independent variable for a given aiccuracy. Trhis considera tion is of particular importance in trajectory simulation work. 2, Winds are obtained from direct measurements (b,,alloon observations, or PKA01Vs, for the lower aitm-osphere and rocket observationsor lROCOl3's, for the uipper atnmosphere), from forecasts based on ohservations~or from l'orecasts based on statistical analyses of many previous observations (climatological data). Thiese sources are listed in the order of decreasing accuracy, and all sources -47-

THICJOHN$ NOPKiNS UNiVItPS'V

APPLIED PHYSICS LABORATORY


SILViE lPIRING MARTVLAND

are frequently employed in some phase of annlysts of body


motion, A law used extensively in the forecast procedure is the geostrophic wind law, which is derived by equating
the Coriolis and pressure forces, comes invalid at low latitudes. However, this law be-

3. If the mo.."txirc content of the air is high, thre perfect gas law n-ust be applied using partial, pressures of air and water mixtures with appropriate gas constants. Sometimes moisture is accounted for by using an effective (or virtual) temperature. The effect of water vapor is to decrease the density when it is compax'ed with a computed density in which the moisture content is neglected. Neglect of the moisture effect results in a density error of a few percent at most, and the error decreases rapidly with decreasing altitude and decreasing temperature. For most analyses this effect may be ignored.

4BI
Ii

- 48-

~tH9

JO~HN$ HOPKINS w~.I~(RSITv

APPL ILD PHYSIC93 LABORAI CRY SMtVIN OPPINU M~qILANV.

ie s; dlivided rinto thi'wee br)oad roltitude T1he ~ m~p ry ridnterplaneta bands cal led tile haro sphiere, uxc ~pher e, :11 uCC1LP to h (xisti froml S(e:ilvl g,18 (tPIg. 3i-1). T110' brosp13)IM.'O eoper lih. about 500 km1 (1. (65 . 10 feet or 275 rim i).
Cxi~tS fV01m 500 to ;1h(oot (GO 000 kmI

e1x0Hphere'L,

theC iij1r ruIit; ever collide with on, ainother.

jso

III the (331 000 Mi110. idlyI klow' tjiot. iton lii thle Molecules enltering

nt

baise of this region dlescribe ballistic trajectories, and some, tra vel beoynd the gin Vi taltLIonal 1'i(-ld of theL planet, III thilik region, temlper~ature -ins rico significnr ice since there is rio Mnxwollalan distribution of vel'ocity. Int er planetary gas u-cists at tittitudos above about GO0 000 km1.I The barosphere is subdivided into the troposphere, stratosphere, niesosipherc, and thermosphere with thle tropopnuse, it~ratopnuse, mresopause, and thlermlopause locnted between theose regions, Lis shown in FPig. 3-1, These altitude bands are, baised upofl thle temperature variation with altItude,. Tile region from sea lev~el to the top of tile mesophere is also called the homosphere. The behavior of each altitude band varios somewhat with cartint latitude, Many types of standard atmospheres are in commion use; only three will be considered in this section. Thle 1062 Standard Atmosphere is typical of middle latitude locations; the Tropical Atmosphere is typical of low latitude locations (01 to 20' N. latitude), and the Polar Atmosphere is typical of high latitude locations ~(north of 60' N. latit~ude). In thle troposphere the variation of temperature with altitude is strongly influenced by the earthi acting as a heat source. The temperature, in general, decreases with increasing altitude. The top of thle troposphere occurs at an altitude of 36 000 feet for th-e Standard Atmosphere, at 55 000 feet for tile Tr'opical Atmosphere, anid at 83 000 feet for the Polar Atmosphere. Essentially all weather phenomena occur in the troposphiere.

* *

APPLIED PH.Y$iCS LA1HORA1CR'v

The.( tropopause is the top of' the troposphere Ind is -an altitude of minimi-umn temperature. i'or the Standu rd Atmosphere the temperatlure is constant from 3(3 000 to 66i 000 feet, where it begins to rise; for the tropicail ntmnosphere the temprnlnture begins to rise rit 55 000 feet; for the polar' atmiosphecre the temiperatureVL is (mnstant to Wi 000 feet, Where it b. girls to rise. The teniperatLre rise is a
result of ;I)5Iovploiili ot iltI8\'i(1i'ht sulni, rtidiations by

ozone. The temper'ature Conitinlues to riisc with mecca sing altitude to about 1.563 000 feet for the Standard Atmosphere. Characteristics to suifficiently high a.ltitudes --re not avaU.aible for the tropical and polar, ati-nosphereSg. Above- the stratosphere the tei-perature again decreases with increasing altitude., reaching ;i mninimum temperature at the inesopause. Pox' the Standard Atmosphere, this 1ltitud.a is 263 09J0 feet. Above the m-esopause the tempera-

ture Again Increases with increasing altitude up to the

therniopause--, which occurs at approximately 1 650 000 feet (275 nmi). Thu temperature rise is caused by the absorption of ultravio~let rays with a wavelength of 100k. The homosphere (sea level to 300 000 feet) is the altitude band of primary interest for analysis of reontry

body motion.

In this altitue region, the atmosphere Is

characterized by constant proportions of N 2 , 021 and argon, and the atmosphere obeys the hydrostatic equilibriumn and perfect gas equations. For analysi-s of reentry vehicles, tile atmospheric proJerty of most interest is air density; of secondary interest is air temperature. The ratios of density f'w the tropical arid polar atmospheres to the density for the 10)62 Standard Atmo)sphere are shown in Fig. 3-2. It may be expected that the tropical. atmosphere would have a lower density and the polir atmosphere a higher den-

'

I
'50

~sity than those of the 1962 Standard Atmosphere.

over, this is true only for altitudes b~elow about 20 000 to 30 000 feet. Therefore, depending upon thQ vehicle design and trajectory, the tropical day may be equivalent

Ho0w-I

to an atmosphere of greater density than tha~t of the 1962

LRW.....

THN JOH04%"OPM04% kINIVIIUIISITV

APPLIED PHYSICS [.ABcM4ATnF1Y

Starviard Dlay, and the polar day may be, equivalent to An atmosphere of lessor density than thnt of the 1062 Standard Da y. The toemperature, in addition to the effect on density, qffects Cie sp.)cd of' sound, and thas Mach number VI/a). rhe spceed of sounfd ( =iWT) is1 proportional to the s qua rc VMut of' tiw nvnibi eat ternipe 1'tu re aad tli o'e c haraneteristics are aLso shown in Fiig. 3-2, Notu that the sp,ud of sound at a given altitude does not vary from. one standard atmosphure to anothev by more than about 7. 5l%. This amount of variation is r'elati~vely unimportant. How-* ever, extreme variations in temperature from the I J62 Standard Atmosphere, such as those of the winter months -it northern latitudes, may have a significant effect on a reentry body trajectory by affecting the Mach number history. Some, characteristics of the 1962 Standard Atmosphere are given as a function of altitude in Table 3-1. More elaborate tables Lire given In R~ef. 3-1. Similar data for the tropical and polar days are given in Reis. 3- and 3,3. 3. 2 Variation of Density and Temperaturc with Altitude Consider an elementary volumne of air (page .52) of base area dxdy and height dh, If the elemnent: of volumo is in static equilibrium (the acceleration of the elume--nt of volume is zero), then the following equation applies: dxdy(P + dp dh) -Pdxdy + W The weight of the volume of air is given by:
:

W = gdxdydh
Therefore, dP
-

gdh

(3-1)

-li1t

A HIE JOHN$ OPKII4 UNIVCRNITV APPLIED PHYSICS LABORATORY


SILVIA

SPRING

MHATLAND

p+dP

R
dh

dx

This equation Is called the hydrostatic equilibrium equation, and all three standard atmospheres dlicussed in this section obey this equation, A second equation that is valid in the homosphere is the perfect gas law which states!
P = RT , (3-2)

Differentiating Eq. (3-2):


dP - R(Tdo + pdT)
.

(3-3)

Substituting Eq. (3-3) into Eq. (3-1) and expressing g in terms of go: p T RTg

By definition:
dh G go0 dh (3-5)

52

......................................................................

THKIJOHPN5 HOPKINN UNIVI"GIty

APPLIED PHYSICS LABORATORY


OILVI1mNPRING MAR.TLANU

where hG is called the geopotential Rltitude and h is the geometric altitude. If g is assumed to vary inversely with the square of the distance from the center of the earth: R
0

('-6)

Substituting Eq. (3-6) into Eq. (3-5) and integrating: hG


-

I
,

+hh

(3-"7 )

or
h h G

1 "-

hG

(3-8)

In the remainder of this section, the solution to Eq. (3-4) will be given in terms of the geopotential altitude. However, the results may be obtained in terms of h by using Eq. (3-8). It should be noted that, since Ro f 2. 09 x 107 feet, h f hG for low altitudes. Even at h 1 100 000 feet, hG and h differ by only 477 feet. In terms of hG, Eq. (3-4) becomes: dT = d_ + p -T go -- dh RT G (3-9)

'4

The closed form solution to Eq. (3-9) depends upon the nature of the variation of T with hG. For practical applications it is necessary to consider only two cases, a constant temperature (isothermal) relation and a linear relation.

- 53 -

hic APPLI rJ

JOHNi

HO4KINO UNIVUEROISY PHYSICS LABORATORY

IILVEE1 S'PlNG M&I.yLA..D

Consider an altitude interval from hGn to hG. grating over this altitude band, Eq. (3-9) becomes:

Inte-

go n n

hG Idh hG n

(3-10)

If T

is constant,

Eq. (3-10) maybe integrated to give:

h
P -"P exp
Gn

-h
(311 13-1,l)j

where RT' H - 90 If T varies linearly over the altitude band, then: +dT T nT+(3-12)

d h--

(h

- h

Gn

(3-13)

In this case, Eq. (3-10) may be integrated to give: R dT ,(3-14)

.2

For calculations, it is frequently desirable to define the density and temperature variations with hG by an equation rather than to use tables, If the atmosphere obeys the hydrostatic equation, it may be divided into altitude bands for which the temperature is accurately defined by linear variations of T with hG. Then the density and temperature within a given band are defined using Eqs. (3-11), (3-13), and (3-14) and the conditions at either the lower or upper boundary of the band.

54 -

A!

,-

..

- ':

. ,. .

.. l'

. ,.

, -,'. .,, . .,,,,C .. ,;: k,., .- : ... . .'.

Ir UNIV1 at PHYSICS HOPKINS T JOHNII APPLIED LABORATORY .vian sPlUme MARYLAND

Iy

The characteristics of the 1962 Standard Atmosphere may be defined to within 0. 5% accuracy using Eqs. (3-11), (3-13), and (3-14) and the data in Table 3.-2. For simple calculations, it is frequently assumed that the entire atmosphere is at a constant tempernture. Such an atmosphere is called an isothermal or exponential atmosphere. In this case, Eq. (3-11) becomes

Values of

and

H are selected to provide the best accu-

racy -for pin the altitude region of most interest for the particular study. The quantity H is called the scale height. From Eq. (3-11), the scale height is the change in altitude corresponding to a change in density (or pressure, since T is constant) by a factor of 1/e. 3.3 Nonstandard Atmosphere

Frequently it is necessary to consider atmospheres other than the Standard Atmospheres. The quantities of primary interest for studies of reentry body motion (neglecting such specialized studies as the effects of wind shear and wind gust) are density, temperature, horizontal wind direction, and horizontal wind speed. These data are available from observations, forecasts, or' climatological characteristics. The latter data are characteristics based on statistical analyses of many observations. Observations are made in two altitude regions. Balloons are used to obtain temperature, pressure, dew point, and wind data at altitudes from the surface up to the highest altitude possible. Frequently, the upper limit in altitude is about 30 000 feet. Sometimes the data extend to observations at about 100 000 feet. These observations are called RAOB's. In the northern hemisphere, RAOB's are obtained daily at OOZ and I 2Z hours at many observation points. Rockets are used to make observations of

55 -

- 1

"-

-U--

?41 JO1N4H HOPKINS WNINY 1RI17YV

A PPLIED PHYSICS LABORATORY


SILVIN SPRING. Mh*YAftCA

temperature, density (sometimes), and winds as a function of geometric altituie for altitudes from about 60 00nf to 200 000 feet. Data from these soundings are called ROCOB's and are made only at a very few locations, usually near 12Z hours Mondays, Wednesdaye, and Fridays. 3. 3. 1 RAOB )at-DryAir The RAOB pressure and temperature datta are converted to density data by Eq. (3-2), The geopotential altitude corresporiding to the observed P and T are calculated using the hydrostatic equation. Since data aro obtained at a large number of altitudes, the temperatures from one measurement to the next may be averaged and used as a constant so that Eq. (3-i1) applies and may be written in the form: Tn+I) h Gn+l uhG hn + R(Tn 2 + gn+n n (3-16)

where n corresponds to the measurement at one altitude and n+1 corresponds to the measurement at tne succeeding altitude, Equation (3-16) is evaluated in stepwise fashion starting at the surface conditions where hGn is the surface elevation of the observation location. 3. 3. 2 RAOB Data - Moist Air The values of density computed uaing Eq. (3-2) are valid provided the effect of water vapor may be neglected. The effect of water vapor on density may be evaluated by writing the equation for the density of the air-water mixture as the sum of the partial densities: PA + "w RT R T
w

(3-17)

I I,

a-.

,1
IN
i

JH i OMl oPHNOkl UNIVIrNSIII l


lilL~S Y111 B.I...l MUY&lli CN

APPLIrD PHYSICS LABORATORY

where P "

A + Pw
0. 622:

In terms of p and using R/Rwfl

Pw 0( 1
-

o 378

.(3-19) ,)

If the moisture content is given in terms of relative humidity, 0 - Pw/Ps, rather than Pw, then,

,'

1 - 0. 378

(3-20)

The water vapor pressure, Pwo is the saturation pressure at TD, and P 5 is the saturation pressure at T. Empirical
equations for the saturation vapor pressures are:

P.in. exp 2. 65

Ps i

exp
5T

6i5

whc2.

g-ne

rThe effect of water vapor is to decrease the dens~ity compared to a computed density which ignores the effect of moisture. given by: The percent difference in density is

37,.
37

o D
exp 22. 65 - 99

(3-22)

Tx

A plot of the error as a function of altitude anid dew point is given in Wig. 3-3. Also shown are the standard

57-

--

T'lE JOHNIS HON INS UNIVERSIYV

APPLIED PHYSICS LABORATORY


il-VIM SPRING MANiNhANG

temperatures for the 1962 Standard Day and the Standard Tropical Day. For these standard days, the effect of 100% relative humidity (TD = T) on percent change in density is maximum at sea level and is about 0. 7% for the 1962 Standard Day and about 1. 8% for the Standard Tropical Day. Only for very hot humid days is water vapor likely to have any significant effect on density, and even this effect is significant only for very low altitudes.

3. 3.3 1OCOB Data


The effect of moisture is negligible at altitudes where ROCOB data are obtained; therefore Eq. (3-2) is valid. When density and temperature both are available from an observation, the only data reduction required is the conversion from geometric altitude to geopotential altitude by Eq. (3-7). However the density data are usually not nieasured directly. In this case, Eq. (3-16) may be used in the following form:r

0n+1

n [R(Tn

+ Tn+l) (hGn - hn+l)

3-23)

where p+l is obtained in stepwise fashion using a known initial value of density (from RAOB data, for example), measured T, and hG computed from Eq. (3-7) using measured 3.3.4 h. Winds

A wind is defined in magnitude (speed) and direction from which it is blowing, measured clockwise from the north. In some current forecast procedures, the predicted winds are not based on measured winds but upon measured pressures and temperature, using the geostrophic wind equations, The geostrophic wind is obtained by balancing the horizontal pressure force acting on an elementary mass

- 58

THR JOI, NI HOPKINS UNIV1PSITV

A4PLIED MiMV1M PHYIICS LABORATORY 60MINi. MARLAP4

of air by the Coriolis force, The Coriolis force is perpendicular to the horizontal velocity of the elementary mass and, in the Northern Hemisphere, is directed to the right (to the left in the Southern Hemisphere). The magnitude of the force acting on an elementary mass of air, pdxdydh, is given by the equation:

Fc

2V wto

sin 0 dxdydh ,

(3-24)

A horizontal pressure force acts on the elementary mass whenever a horizontal pressure gradient exists. Consider the following horizontal high pressure system. Let x be the horizontal distance perpendicular to the isobars, y the horizontal distance parallel to the isobars, and h the vertical dimension,

Vw

ISOBAR

Fi

FC
X

HIGH
PRESSURE

dx

The pressure on the two surfaces of area dxdh are equal, so that the net pressure force acting on the mass in a direction parallel to the isobar is zero. The pressure on the two dxdy surfaces are unequal, but the weight of air is just balanced by the pressure force (hydrostatic

59

?He JOWNG

HOPKINS

UNi~aRSITV

APPLIED PHY4SICS LABORATORY


SILVIA $PRIM*S MAPVLAN13

equilibrium is assumed), so that no net vertical force exists. Let the pressure acting on the dydh surface that L~s farthest from the high pressure center be P. Then the pressure oui the dydh surface nearest the center is Fr + aF dx. The corresponding forces are Pdydh and (P + dx)dydh. F u~ Therefore, the net force is,

dxdydh,(2)

Isphere),

The total horizontal pressure force is perpendicular to the isobar and directed away from the high pressure, Since the geostrophic wind is obtained by a balance of the pressure and Coriolis forces, the Coriolis force is perpendicular to the isolbar and acts toward the high pressure center, am ehown in the sketch, For a Coriolis force In this direction, the wind maust be directed along the isobar with high pressure to the right (for the Northerm Hemi-

Sinope F. c Fp 5 from Eqs. (3-24) and (3-25):


2V siW (3-26)

Equation (3-26), the geostrophic wind equation, Is a,curate except for the first few thousand feet above the surface where friction effects cause the wind to be inclined away from the Isobar toward low pressure and the magnitude is lower than that given by Eq. (3-26), Typically, the mnaximum deviation in direction is 250, and the maximum devia. tIon in speed is a factor of one-third to one-*half the geostrophic wind, depending upon surface roughness. In addition, the equation is not valid at latitudes near the equator, The equation is valid for the Southern Hemisphere, but the direction of the velocity along the isobar is opposite to that discussed for the Northern Hemisphere,N Equation (3-26) is applied at various altitude levels to obtair. the variation of wind speed with altitude. The

f.

~~6

IfIlbulll

?me JO4N$ "OKINI UNIVgNSI1'Y APPLIED PHYSICS LABORATORY


SIDII MARYLANII

wind direction is obtained from the slope of the isobar at the location of interest. Therefore, accuracy in wind characteristics ,'equires accurate estimates of density, horizontal pressure gradient, and shape of the isobars.

tV

.i.

-51.1

,'-

1!
T4e JOMNS 400IWO uI1N'1I0I6TV

APPLIAO PHYSICS LMORATORY


SILVIA bFA'N6 M&AIM~ANf

REFERENCES

"3-1

U. S, Standard Atmosphere,

19G2, prepared under

sponsorshlp of National Aeronautics and Spoce Adminiistration, United States Air Force, and

United States Weather Bureau, Washington, D. C., December 1962, 3-2 3-3
-

Climatic Extremes for Militar.j_,utpment, Std. 210A, 2 August 1967,

Mil.

L. F. Fehlner and E. V. Nice, Tabulation of Standard Atmospheres at 100-Foot Intervaos of


A1lt~ude, APL/JItU TO313-1; AugM 1058.

PI

I
K

1'
__ _________ --. 4 .,.%i1 -

;I
T1,9 JOHN% HOPKINs UNIVtP5Ily

o;APIILf PHYSICS LABORATORY


SILVIO S*poIh MARVI.AND

INTERPLANE-rARY

GAS

000
NOTE: NOT TO SCALE

FEET33
EXOSPHIFR,

000 rirnr

ABOVIE 320 0HEFEETLBEnm

l
320

--

'r~HE=RMVOP'AUSE

tl

275 nrmi
i

T"HEAMOSPHEE -/MESOPAUSE--

240-

200 -

--

STRATOPAUSE--

160

120
STRATOSPHERE

40 ----

TROPOPAUSE

" TROPOSPHERE

3W0

340

380

430

480

500

!40

TEMPERATURE ( RI

Fig, 3.1

TYPICAL ATMOSPHERIC TEMPERATURE PROFILE AND ALTITUDE REGIONS

Preceding page blank


V~

TM9

JOHNS NOPRIN48 UNIVWKNSIY

APPLIED PHYSICS LABORATORY

06

L:iI
I-

ccJ

LL.

-66 -

THE JOWI.S MHOPKINS UNIvUlI?v

APPLIED PHYSICS LASOAtOmY

0,5
70
-

1.0

2.0

3.0 4.0 5,0


-

--

- .. .

. .

e- 0.1%
50

:40

S30

STAi",ARD

TROPICAL TE
20

PERATUR
,

1962 STANDARDTEMPERA

10N

o
-60

,I
.40

,
.20 0

,I,
20 40

,\i
60 so 100 120

i.

DEW POINT (0 F)

Fig, 3.3

EFFECT OF WATER VAPOR ON DENSITY

-67-

Tilt JOH4N%9OPKII4S UNVIEtNSITY

APPLIED PHYSICS LABORATO~RY


SIVI SPRING MAQVI.ANfl

in -L0ca
IS

m ccI

(0

M0 f(0 M0 4.tr
t

- in

In

al

1 0 if

-~lMC M

"II
('I CV M 0 -to 1 o)
4

4'4

M-

to-e

-zt

+- +--

+++

t,

C;. ic

6cl

NN L0 0)
ell coc mC -

~
In

o-

-4

-4

-4 I

W ca a.0(0 to M..t4N..i-.4.-~ m

.. i.. M

NI

'10 44 0

r'C4

~in

M)

tHol"0in

04 WInN

-4.

-.

00

-u

I
fWI JOINS 4OPRINS UNIVBPSIBT

APPLIED PHYSICS LABORATORY


SI1VIN SPRIN'k MA*VLA.D

Table 3-2 Constants for the1962 Standard Atmosphere

dT Td
(feet) 0.000 36089.238 65616. 750 104985. 875 154199. 375 170603. 625 200131.188 2591B6. 313 291153. 188 323002. 625 354753. 250 386406.125 480780. 813 512045. 938 543215, 188 605268. 375 728243. 750 939894. 688 1234645. 000 1520799. 000 1798726, 000 2068776. 000 (0 R) 518. 670 389,970 389. 970 411. 570 487, 170 487. 170 454. 770 325. 170 325. 170 379. 170 469. 170 649. 170 1729. 170 1999. 170 2179.170 2431. 170 2791. 170 3295. 170 38B9.170 4357. 168 4663.168 4861.168 (fR/ft)

(slug]ft3) 2, 37680982E-03 7. 06119929E-04 1. 7082000OE-04 2. 56609928E-05 2. 769 79 972 E-06 1. 47349965E-06 4.871 89993E-07 3. 88259984E-08 6. 1507989BE-09 9. 65109992E-10 1. 907099B)E-10 4, 3, 2. 1. 8. 3. 6. 1. 3. 9. 3. 72659967E-1I 56239951E-12 24879924E-12 55919982E-12 43449DN4E-13 03459948Eh-13 95599719E1-14 26079973E-14 05989996Z-15 00309B37E-16 04629894E-16

-3, 56619991E-0:3 0, OOOOOOOOE+00 5. 48639B00E-04 1. 536199B2E-03 0. OOOOOOOOE+00 -1, 09729986E--03 -2. 19459995E-03 0. OOOOOOOOE+00 1. 69529999E-03 2. 83429981 E-03 5. 68669662E-03 1. 14439987E-02 8. 63499939E-03 5. 77439740E-03 4. 06099856E-03 2. 92699994E-03 2. 38109985E-03 2. 01519998E-03 1. 63550000E-03 1. 10109989E-03 7. 32979970E-04 0. OOOOOOOOE+00

'The temperature gradient is applicable to the altitude hand which the gradient is listed and for altitude at the beginning altitude. highest next ending at the

I'

,-69-

.....*~.. --- ,,-. ... ......................................

THE JOHNSHOPKtINS UNIVERSITY

APPLIED PHYSICS LABORATORY


6ILYRR SPRING6.MANYLAND

4, EQUATIONS OF MOTION

Precdingpageblan 71

21
TI JOHNS HOPKINS UNIVRIrIIIITY BIN-VIP SPRING. MAVL6AND

APPLIID PHYSICS LABORATORY

The motion of a reentry body during ballistic flight is affected primarily by five factors: 1. Gravity,

2.
3. 4. 5,

Earth rotation,
Atmosphere, Body dynamics, Asymmetries,

Each factor is considered in simplified form to show the general effects on the trajectory.

',.,

73.

Preceding page blank

I I I I I

YNI J@44NM@9WN1UNIVSNUIYV

APPL ID PHYuICS LAIOATOY m&wse I.m. MAN'i.ANS

4,1 EFFECTS OF GRAVITY

I I
WA

I
4*1
I.

t 5 L #1

Preceding page blank

-75-

Fr

?HI JOHN$ HOPKINS UNI1[eI?.y


I ,"

APPLIED PHYSICS LABORATORY llSIVIN


SPillO
MAIWILhI.D,

A. ,

SYMBOLS FOR SECTION 4.1 Symbol g I K R r r t V V Definition acceleration of gravity constant defined by Eq. (4.1-15) constant defined by Eq. (4. 1-5) range distance from the center of the earth to a point on the trajectory r/r time velocity /VI -r 0 path angle central angle, measured from apogee degrees radians or degrees seconds ft/see feet or nmi feet Typical Units ft/sec2

Sflight
6

Subscripts a e o refers to conditions at apogee refers to conditions on the surface of the earth refers to initial conditions

-77

Preceding page blank

4.....4 ... ....

ThE JOHNS ,OPKINS UNIvIIITV

APPLIED PHYSICS LABORATORY


MILVE
SPRINdS MARYLA0d1

SUMMARY AND CONCLUSIONS

The effect of gravity only on the trajectory of a ballistic missile is considered. Equations that define the trajectory (velocity, time, flight path angle, altitude, nnd range) in terms of initial conditions are presented, Plots are given which show the relationship of range and flight time as a function of initial velocity and flight path angle for minimum and nonminimum energy trajectories. Also shown are plots of range sensitivity to initial velocity and flight path angles, An example of the uses of the equations is given in Example 1, Section 4.6. The major conclusions are: 1. A ballistic missile follows an elliptical flight

path.

For a given initial velocity, there is an initial

flight path angle that results in maximum range, and the trajectory is called a minimum energy trajectory. For minimum energy trajectories, the flight path angle decreases from 450 for very short range to approximately 320 for a range of 3000 nmi; the velocity increases from 0 ft/sec at zero range to approximately 20 000 ft/sec at 3000-nmi range; the total flight time increases from 0 seconds at zero range to about 1300 seconds for 3000-timi range; the trajectory apogee increases from 0 nmi at 0 nmi range to about 550 nmi at 3000-nmi range (a maximum apogee of about 700 nmi occurs for a range of 5500 nmi and a flight path angle of 22. 5O). 2. Trajectories for which the initial flight path angles are greater than those for the minimum energy trajectory are called lofted trajectories; trajectories for which the initial flight path angles are less than those for

the minimum energy trajectory are called delofted, or


depressed, trajectories. For a given range, the total flight time increases with Increasing flight path angle.

Preceding page blank

weINSP4

HON OPKINS UNIVUNUITY

APPLIED PHYSICS LAUORATORY


BILVUR UPPIN MAm1mIaNIS

3. The range for a minimum energy trajectory is extremely sensitive to initial velocity (about 2800 ft/ft/sec at 3000-nmi range) but is very insensitive to initial flight path angle. With Increasing loft at a range of 3000 nmi, the sensitivity to initial velocity decreases but the sensitivity to Initial flight path angle increases, For delofted trajectories at 3000-nmi range, the sensitivity to both initial velocity and initial flight path angle is greater than that for a minimum energy trajectory,

II

ii.

80

SI* .

Tl~ JOHIN@ MOOKINN UINIVR"ly

APPLIED PHYSICS LABORATORY


GILVEISSOpINSMAVLA&ND

"Consider a nonrotating, spherical earth without an


atmosphere. the sketch. The geometry of the trajectory is shown in

VO
0V

The trajectory of a body with initial velocity Vo at ro and fired with a flight pathyo may be obtained by applyIng the laws of gravity, conservation of energy, and conservation of angular momentum as defined by Eqs. (4. 1 -1), (4, 1-2), and (4,1-3), respectively. goro 2 Vo 2
2

gr 2 V2 t

(4. 1-1)

~
0

gr g
0 0

gr

(4.1-2) (4. 1-3)

r V cos0
00

rV cos

V2 for

It may be shown that the trajectory is elliptical V2

2g r0' < 1, parabolic for 2g--r


0 0

1, and hyperbolic for

0 0

-81-

.I

THE JOHN$ HOPKINS UNIVENS8tY

APPLIED PHYS!CI LABORATORY


SILVIR SPuINs MASVLAND

V2
2goro >
00 0 0

V2
0 1. For ballistic vehicles, 2goro, 0 is always less
0 0

than I so that the trajectory is elliptical. Using Eqs. (4. 1-1) through (4. 1-3) and the properties of an ellipse, it may be shown that the trajectory is defined by the equation:

- 2 cos2
0 I 0- K cose 8,

!
(4.-4)

where

r Kr i

.
0

-2

K - I - VE
equation:

14

byo sti

(4.-5)I

The range is the arc length OB and is given by the R 2roe (4.1-6)

.,
'

where 80 can be obtained from Eq. (4. 1-4) by setting

I and solving for cos eo.


I - V-

Thus:
!

Cos 19 =

Cos2_ o, . . .

or
"

tan

V0 cov 0 sin 7 0 -2 C 2 1V 2cos8O2


I-V
0 0

'N
(4. 1-7)

"-82 -

ii

?"t JOHNS HOPKINS UNIVISII

'

APPLIE D PHYSICS LABORATORY

It should be noted that the equations derived above are general insofar as the reference nltitude, re, Is (-oncerned, That is, ro is not necessarily a surface coridition. H-lowever, if rv is not n surface condition, care must be exorcised in evnlunting the icference vLlodty, V-79g 0, and It . In terms of surface conditions:

'r

(r 0 0 )

, 26 000
0 O0

(4. 1-8)
(4-n

2r.6 R~ e

,0

From Eqs. (4, 1 -6) and (4, 1-7), it is observed that a given range may be obtained for varioua combinations of However, there is one combination of Vo atid Vo andy , *o that resuilts in minimum Vo. This trajectory, called the minimum energy trajectory (MET), represents the minimum booster impulse forx a given range or, alternatively, represents the maximum range that may be ob-

tained from a given booster.

The minimum energy tr,-

Jectory is obtained by operating on Eq. (4. 1.-7) using standard procedures for obtaining maximum or minimum values, The results are the following:

yo~ :2+ "-"tr

"-o} o (4. 1-10)


T3

or * y0
-2
V0

70=45 -T+:
-tan

min

(4. 1-11)

It is noted thnt '/ decreases from 45' at very short range to 00 at maximum range (R/ro ft). The velocity -gr. at maxinmum increases from 0 st zero range to Vo range. The latter velocity is the velocity for a circular

L
!-+

[I

-83-

?11JOWNI NO NINOS ONIV906tV~1

APPLIED PHYSICB LABORATCORIY lSvIm 3pmlh,2 kAII?&.AND b

orbit. For ranges that will be considered in this report (R1 rr.Jqs. 1000 to 3000 nmi) the values of y and V0 computed from 4-2. For 4-1 and in Figs. (4, 1-6) and (4.1-7) are shown
all examples given in this section, ro
=
2 re

2.0 9 x 107

feet,
The time of flight may be derived in the following
manner. The rate of change in 6 expression: is given by the f'ollowing

Cos
r

(4,1-12)

:
u

The negative sign is required since, for convenience in deriving the equation of the ellipse, 8 is measured from apogee, Using Eqs, (4. 1-3) and (4, 1-12) the equation for time, becomes:

r
t
VCos

r d0

(4,1-13)

Substituting for r from Eq, (4, 1 -4) and integrating gives the result:
ts

(V 0 cos '0)

(4.1-14)

where I

is given by the equation:

In I
Vo
2

tan-' Ko

- Va

Cosn o,

- -I (9/ 2 )\

(4.]1 -15)

'

I -K

and is evaluated between the limits

e and So,

If one is interested in just the total trajectory time, Eqs. (4. 1-14) and (4. 1-15) may be simplified by noting - 84 -

'Ii
I I t !il

'

ll

l ' l......l l,

l'

TM9 JOHNS HOPKINS WNIV3R1ITV

APPLIED PHYSICS LABORA'ORY


SILVt SPlRING MARYLANO

that the trajectory is symmetrical about E equation for the time becomes:

0, and the

2V g

Co
...

K sLn8
+ ta

j -1K, tan (6c,/2)


K (4.1-16)

A plot of Eq. (4, 1-16) is shown for a range of 3000

nmi in Fig. 4-3. For a given range, time increases with increasing initial loft angle, -/,. The minimum time for a
range from 0 to B is that computed for a circular orbit trajectory from 0 to B; that is yo = 0, Vo - gC-ro, and t 692 seconds for R = 3000 nmi. For a minimum

energy trajectory, K written as!

tanyo and Eq. (4. 1-16) maybe

tar
t 2 1J go -2

tar -&
0

Cos 3

[tan a y2

"

ol

tan'

1/-n)o ta~n V'

(4.1-17)

The time for a minimum energy trajectory varies


from 0 seconds at zero range to 1rYTO7"- = 2520 seconds for maximum range (R = ffr 0 ). For ranges of 1000 to 3000 nmi, the flight time is shown in F!g. 4-4. The local flight path angle may be expressed in the form:

dr
tan " (4.1-18)

I6

""8.-..

?HL JOH4N$ HOPKINS UNIVIRI[ITY

APPLIED PHYSICS LABORATORY


SILvE1

SPINGl

MARYLANO

Solving for dr/d8 from Eq. (4. 1 -4) and substituting into
Eq. (4. 1-1 B):

tan

= 1-

K sin 84l1 K cos 9

Differentiating Eq. (4. 1-19) with respect to t and


using Eq. (4. 1-1 2), it may be shown that j is given by

the following equation: 3 K(K - cos 9) V cos (4.1-20)

(1 - K cos 6)2

r .

Evaluating Eq. (4. 1-20) for ccnditions near impact, it may be shown that # is very small for 1000 to 3000 nmi
trajectories. Therefore, in the absence of an atmosphere, the flight path during reentry is nearly a straight line. This fact will be useful. later when we consider the effects of the atmosphere. The velocity at any point along the trajectory may

*1

be obtained by solving Eqs. (4. 1-1), (4. 1-2), and (4, 1-4) to obtain the following equation:
-=

V
V

2K cos 8 + 1' cosy


=y
0

(4,.1-21)

(4.1-21):

At apogee 6 -

0, and from Eqs. (4.1-4) and

-- VCos 2 27
00

I -K

SV V

I -K cosy

"8

-86-

TH4JOHN$
SILVSP

HOPKINS UNIVEREITY
SP~ilNO MARYLA~N

A4PLIED PH'ISICS LABORATORY

orl~t.
o-

For a minimum energy trajectory, K = tan yo and

-=1 - tan 2 y0, so that the altitude (r - ro) and velocity

at apogee are given by the equations: h V =-

(si-n

+ cos ) 0 ) cos 'Yoi +tan 1o 1 +tan )I

1,

(4.1-22) (4.1-23)

" Cos
0

or

"a o

cos ^/ + siny

(471-24)

For a minimum energy trajectory, the altitude at apogee is maximum (ha %; 700 nmi for ro = re) for Y. 22. 50 (R o 5500 nmi) and varies from about 250 nmi for R = 1000 nmi to about 550 nml for R = 3000 nmi. Since the sensible atmosphere extends to about 50 nmi, most
of the trajectory occurs in the exoatmosphere.

The ratio

of velocity at apogee to initial velocity decreases from 1 at zero range to 1/V-for maximum range. For ranges of 1000 to 3000 nmi, Va/Vo varies from 0. 71 to 0.72. Another consideration of interest is the sensitivity of range to the initial values of VD and 7 0 . These sensitivity values may be obtained in the following manner:

Wvo ae 0 av0

(4.1-25)

aRi/6.6 0 and~60 /av may be obtained fromn Eqs. (4. 1-6) and (4. 1-'t) so that:

24)0 2 0o

(F

V sin 2oy K 2R K

(4.1-26) ,

-87II'

1i'il JOHN& HOPKINSI UNIVIIAITY

APP LIED PHYSICS LABORATORY


SI.VglR@PRNQ MANyRLAN.

Similarly, aR/ayo mrnybe obtained from Eq. (4.1-7): -2 2r S


5 7

V
3

(cos 2Y

-- 2 2 -V cos K2

vY)
(4.1-27)

T.

Plots of Eqs. (4. 1-26) and (4. 1-27) are shown in Figs. 4-5 and 4-6, respectively. Note that the range of a minimum energy trajectory is sensitive to initial speed but not to initial flight path angle, With increasing loft angle, the range sensitivity to Vo decreases but range sensitivity to Yo increases.

-88-

Twi JOwNi Pi~0pKlIlg UNIV5UiII"

APPLIED PHYSICS LABORATORY

"
3600 --p
p

~I

300
450 600

3200

2800

2400

12000

1600

1200

800

400
10

I 12

,[ 14

I
16

I
18

I___ 20

__

___

_ ..

22

24

26

V0 (thousands of ft/tao)

FIg.4-1

RANGE VERSUS 7'0 AND Vo

-89
qI

yma

uNftlvilTv JOiqNBMlOWIN@
51I.6VINpmINO. MARYLAND

APPLILD PHY04CS LABORATORY

70

1O0nnil

1500

2000

2500

3000

S650

--

40 -'

MINIMUM

40

*ENERGY

10

12

14

16

le

20

22

24

26

Vo (thousondi of ft/iecl

Fig. 4.2

INITIAL 7'o VERSUS V, AND RANGE

-90-

"

?MR JOHNS HOPKINS UNIVRRSITY

APPLIED PHYSICS LA6ORATORY


GI.V*Pl OPm.IN. MAETLAND

:1I

3000___I
2600I

1200

1800

1400 MINIMULM ENERGY 1000 30

~
40

_
so 70

FIg. 4.3

FLIGHT TIME VERSUS Y,, R

3000 NMI

91-

THE JOHN$ HOPHINg UNIVERSITY

APPLIED PHYSICS LABORATORY


l. Vi SPRING

MAIYLA.D

IS
_ wU

LJ. 9.

'Iuom

I.

A PPLIED PYISLABOCRATORY

II 1

I>

-~,LL

/c

\. ,/93

?MI JOiHNS HOPKINS LJNIVURSI?

APPLIED PHYSICS LABfORATOHY


MAITt.kNb INILVINROWING.

I-

I-

'il.

/
.--

22 ..
-944

ww

2
1LL.
o o

i9
i __ ... L_ L

..J Ik

UNIVKRGIY O *@PINS APPLIID PHYSICS U8LASOATRY ThU ""lN


uIL6VMN SPRtING. MAPMYAND

4.IFET FERHRTTO

I9

?T| JOHINS HOPKINS UNIVSNrVIY APPLIED PHYSICS LABORATORY


SILVIA SPRING MARYLAND

SYMBOLS FOR SECTION 4, 2

Symbol CA h I L R r r S t e=

Definition axial force coefficient altitude refers to the impact location refers to the launch location range from L to I re + h earth radius = 2.09 x 10x 3441 reference area total flight time it increment in reentry time

Typical Units

feet or nmL

nmi feet feet nmi ft 2 seconds seconds ft/sec pounds W 2lb/f"

V W

velocity in inertial coordinates vehicle weight vehicle ballistic coefficient,

flight path angle in inertial coordinates (positive for launch angles) latitude, positive for N. latitude longitude, positive for E. longitude earth rotation angle during time t azimuth angle, positive direction is

degrees degrees degrees degrees

0
A AA $

clockwise from north


azimuth angle from L to I (Eq. (4. 2-19)) 60 half the range angle for nonrotating earth (Eq. (4. 1-7))

degrees
degrees radian

Preceding page blank

- 9'7 -

. !

UNIV3NSItT 14OP11N0 'rNt JOHNSm


S~ipIo hMARYLANDI ILYII LABORATORY PHYSIC& APPLIED

Symbol Se0

Definition half the range angle for rotating earth (Eq. (4, 2-17)) earth rotation rate = 0. 729 x 1

Typical Units radians rad/sec

10~~-4rase

Subscripts e I L refers to quantities measured relative to rotating earth refers to impact condition refers to launch condition

II
I.I
...

N
.

-98 -

tHRM JOHN$ HOPKINS UNlI'VlRSITV

APPLIED PHYSICS LABORATORY

SLnVIER SPIPNO

MARVLLNI

SUMMARY AND CONCLUSIONS

The equations of motion given in the preceding section ire acpplic('ble to the rotating earth case provided the trajectory is dlefined in the inertial axis system. There.fore, initial conditions must be computed taking into (-onsideration the velocity of the rotating earth. In addition, the amount of earth rotation during the tine interval from launch to impact must be accounted for. In this section equations are given tor evaluating the magnitude of these effects and an example is given in Section 4, 6, Example 2. The major conclusions arem I. Compared to a nonrotating earth calculation, the effect of a rotating earth on a missile fired eastward is to increase the initial inertial velocity and to decrease the initial flight path angle, Westward firings result in a decreased velocity and an increased flight path angle. The largest change in velocity (about 1500 ft/sec) occurs for launches at the equator, for eastward or westward firings, and for very low flight path angles. 2. Compared to a nonrotating earth calculation, the effect of a rotating earth on a missile fired eastward Is to increase the range and flight time. The impact latitude is the same for both calculations. For a westward firing the range and flight time are decreased.

" jj

.I A

99
-fhI -nS09 -

?HIK JOkNBl

IO~MUI*I UNIV([RUIT'

APPLIED PHYSICS LABORA'ORY


*1W*l pPlINO MlAEIL*Pif

The three basic laws used in Section 4. 1 to define the equations of motion are valid for a coordinate system fixed in space (nonrotating earth). Two major effects of the earth's rotation on the flight of ballistic vehicles will be considered. 1. The launch site has a welocity caused by the earth's rotation so that launch velocity and launch direction are modified compared to the nonrotating earth solu-

tion.
2. The earth rotates while the missile is in flight, thereby affecting the location of the Impact point. Assume that a booster imparts to a vehicle a velocity Ve at an azimuth angle Oe and . flight path angle ye, where all three quantities are measured relative to a rotating earth. The earth geometry required for this study is shown in the sketch, page 102.
A

The vehicle will travel in an orbital (or trajectory) plane that contains the velocity vector in inertial space, V, and the center of the earth. The characteristics of the trajectory are defined by the equations derived in Section 4. 1, provided inertial quantities are used. The inertial quantities may be obtained from earth reference quantities as shown below. The velocity vector is divided into three components: Radial component North component East component V sin y - V cos y cos V cos y sin
i

.J

(4.2--1) (4.2-2) (4. 2-3)

Preceding page blank


S-

101

10

1
'U :"", "' ''

.,

1li

JOINB HOpImNo UNIVIRIII"y IILVtO


8PRINO M&*YLAND

APPLIED PI-IY6ICb LABORATORY

0.De"0729 X 10,4 rad/sac

MERIDIAN.

AL

WEST

EAST

li

Af EQUATOR A

r _2.09 x 10O7 fo oat'


3441 mml

I/
, SOUTH

J .

In inertial coordinates the radial component and north component of velocity are the same as those measured in the earth fixed axes, but the east component is increased by we r cos 0. Therefore, the components of the inertial velocity, V, are: (4. 2-4) Radial component = V sin Y e e North component = Ve cos East component = Ve cos e cos Ie e sin (4.2-5)

e + Wer cos 0 (4. 2-6)

-102I''

APPLIED PHYSICS LABORATORtY


S6.6vif SPRING k4ann~AND

THS JOHN$

HOPKINS

uNI~eftrrv

The azimuth angle, flight path angle, and velocity i~n inertial coordinates may be obtained from~ the following eqinatiuns: tn =eastward component of velocity tan orthard ompoentof velocity tan~ tan 1+ ~
w6 o

V Cosy a/in 0 e e e

(4. 2-7)

yradial component of velocity si total velocity sinVV sinl~ V 4 2-8)

~.[l+2cosBY
e

sinl

~Ler~
__ e __

+ (we/ cs
a

(4. 2-9)

For launch velocities of interest, the velocity ratio is given by the approximation: V~ jV=+ cosY I/ si ee we 0. 729 x 10 r cos0 e Ve (4.2-1l) (4.2-12) (4.2-13) (4.2-10)

-~rad/sec

r e-2. 09 X 10'7 feet era = 1520 ft/sec.

From Eqs. (4. 2-8) and (4. 2-10), it is observed that 0* to 1800, V/Ve > 1 and tends for eastward launches (j to be maximum for launcties from the equator (00)for

j li

-103

7I4t JOHN$ HOPKINO UNIV|INI!y

APPLIED PHYSICS LABORATORY


O-LY2I SPRING MARYLANPO

Tninimum energy trajectories (minimum Ve), for direct east firings (Pe = 90), and for low flight path angles (le-. 0). For eastward trajectories the inertial velocity is increased and the flight path angle is decreased (compared to the values for a nonrotating earth); for westward firings, V is decreased and - increased (compared to the values for a nonrotating earth). The maximum variation of V/Ve from unity is +er/V . For V z 10 000 ft/sec the maximum variation in VfVe is about 15%1, Given values of Ve and YeI the inertial values of and Y may be obtained from Eqs. (4. 2-8) and (4. 2-10).

in Section 4. 1. However, the range computed by Eq. (4. 1 -9) corresponds to the range measured along a sphere fixed in space. The range measured on the rotating earth (in the absence of an atmosphere) may be computed as shown below, The orbital plane defined by V and the center of the earth remains fixed in space. During the flight time the earth rotates through an angle:

These values of V and V may be used in equations given

AA = 57.3 w e t

(4. 2-14)

For given Inertial launch conditions, the impact latitu-1e is unaffected by we and, from spherical trigonometry,

sin 0,

[sin 260 cos 0L Cos

L + cos 2o sin 0L1 (4.2-15)

where is the value computed from Eq. (4. 1-7) for Yo0 and V. = V. The impact longitude is given by:, AL - 57.
3

0"et + sin"

o-

(4.2-16)

where t is the flight time computed from Eqs. (4. 1-7) and (4.1-16) for Vo Vando --Y,

-104-

t
I.

T"7 J
l

140PRINIS UNIVIK4I0TY 4NNI


$WIiNG MAYI.AND

APPLIED PHYSICS LASORATORY


SILVIN

01, and A, the range may be comGiven 0 L, A puted from the equaton:,
R = 2re9 e e (4, 2-17)

From the spherical geometry shown in the sketch, the range angle 2 6 e may be obtained from, o 2 cos 01 cos 0 cos A+ sin 0 sin
0

L '

(4.2-18)

*1 .1

GREAT CIRCLE ARC THROUGH L AND I

GREAT CIRCLE ARC ITHROUGH A AND

MERIDIAN EQUATOR A

AA

The bearing of the impact from the launch Atte may be obtained from cosA co Cos cos 2e cos

e
COs SLI sin

26

e sifL

(4.2-19)

S-

105-

105i

Ti

JOHaks HOPKINS UNIVIII11IY"


*ILVI SPRING. MNARVYANO

APPLISD PHYSICS LABORATORY

For

- 0, Eq. (4. 2-19) becomes:

sin 0I
Cos
_ _

(4. 2-20)

-COS

CBAA Cos

Id
Sometimes V, /, and lare given and it is desirable to obtain the corresponding quantities relative to a rotating earth. The required equations are:

v 2 Ve =[Ii _2er
Ye

cosyV sil V

(~ + .WO

Cos 2 0

11/2
(4.2-21) (4. 2-22)

sin Y

V
WeV cos 0-

e
tan er

tan

V cos Y sin

(4. 2-23)

16

'

ij -J
i

APPLIED PHYSICS LABORATORY


SILVIO @"IN*, MANVLAtND 7

I I I

~~~.

ATMOSPHERE EFFECTS OFIGMON

I I
13
I - 107

'

UNIVK90?TV JOHN$ HOPOKIN THR APPLIED PHYSICS LABORATORY


6Vll MANYLAND SPRINGqe

SYMBOLS FOR SECTION 4.3 Symbol


A CA

Definition
axial force speed of sound axial force coefficient

Typical Units
pounds ft/sec

Cmq CN CN D d
g gX

pitch damping coefficient normal force coefficient dC /d& damping factor (Eq. (4. 5-20)) reference length
acceleration of gravity longitudinal acceleration

feet
ft/sec2 g

H
h h max

scale height = RT/g


altitude above sea level altitude at V max

feet
feet feet

" I (K) In
1 2 (k) I

(K/2)n n.l
(2K)n O 1.3.5...(2n+l)2 n,,0 slug-ft
-

nol
e -K 0

pitch (or yaw) moment of inertia parameter defined by Eq. (4. 3-6) constant defined by Eq. (4. 3-26) reference length Mach number - V/a vehicle mass atmospheric pressure

K k1 L M m P

feet slugs
lb/ft 2

page blank Preceding !II

-109-

TMI", JUMNS MOPKINS


SILVIA SPRING

UINIVKRIITV

APPLIED PHYSICS LABORATORY


MAPYV,.AN

,9ymbol
q qs

Definition
dynamic pressure :

rypicai Units

Z PM 2

lb/ft2

stagnation heating rate per unit area total heat per unit area absorbed at the stagnation point gas constant s 1716 Reynolds number, VL/v vehicle nose radius reference area atmospheric temperature time intervals of time defined by Eqs. (4. 3-20) and (4. 3-21) vehicle velocity vehicle weight distance from vehicle nose to the c. p. distance from vehicle nose to the cg. X

T3tu/ft 2 sec Btu/ft ft 2/sec2 L, JR

Q
R 1k RN S T t At1 , At 2 V W X C. p. X c.g.

feet ft 2 OR seconds

2I

seconds ft/sec pounds feet feet feet

4X

-x

[3ballistic
y
V

c.g. c.p. vehicle angle of attack coefficient A

radians lb/ft

reentry flight path angle (always negative) kinematic viscosity atmospheric density

degrees ft /sec slug/ft 3 rad/sec

WA

aerodynamic pitch frequency

-110-

I
T4I9 JMNNS HOPKINS UNIVrADITY

Sl~~it.vS ~ii

APPLIED PHYSICS LABORATORY mil QN A I.AKKibl

Subscpts. ave E 0 refers to average over the time interval refers to reentry conditions (K=O)
refers to initial conditions

refers to sea level conditions

max

refers to maximum value except as noted

A dot over a symbol means the first derivative with respect to time.

.1
'I

I-.-

II
*1

~-

111-

.....................................

*".-..-,n

.i.

iNt JOKNO MOPKWIN UNIVfIItINB

APPLIED PHYSICS LABORATORY


tIiLVf

SpPtNQ M~mA~N~h

SUMMARY

vehicle wovw s throlug(h nn rintm( . phiree, aerodynajmic for'ces nt. genorflte( thAt ;1il'ced. Velicilue notien, The magnitude ul thius Iu U'LUB is L11lUCnlCUd strongly by the missile velocity and atmospheric density. In this section approxi-nate equations nre derived fo;: missile velocity, ambient density (or pressure and temperature), and related parameters.
When i

Assuming that the axial force is the only force acting on the missile, and making other appropriate simplifications, equations are derived tor velority, time, dynamic pressure, longitudinal deculeration, Reynolds number, stagnation point heating rate, and the total heat absorbed at the stagnation point as functions of a parnmeter (K) which, for a given missile, atmosphere, and trajectory, is defined by altitude only. Most of these quantities increase from zero at reentry, reach a maximum value,

A*1

and then decrease prior to impact. Equations areoccur. given which they for the maxima and the value of K at An example illustrating the use of these equations is given in Section 4. 6, lExample 4.

|!Preceding page blank

-13-

4
I l4~k

rii

JOINM,

MOffINB UNI~t~gtY

PHYSICS LABORATORY APPLIED


SILVIR SPAIN* MARYLANDI

The boot;t phase of the trajectory is guided to such a high altitude that atmospheric effects during the exit phase are usually negligible, Therefore, this discussion is limited to the reentry phase. Custo.narily, reentry into the sensiNl ntmosphere is assumed to begin at an altitude of 400 000 feet. The velocity increases slightly under the effect of gravity, reaches a maximum value, and then decreases under the influence of aerodynamic drag, The maximum velocity occurs when the axial force is just equal to the component of gravity along the flight path, or:
C q S W sin y

Using the definitions of q and,, this equation becomes:

AI
h)
VE

max If the reentry velocity, VE, is given for some altitude above that corresponding to Vmax, the maximum velocity may be estimated using the equation: i"
Vmax -VE + k

(h~E " max),i


(h

where k is a constant having a magnitude between 0 (for ) A precise value of hE f hmax) and I (for hE ",> h k is not important since the Jflrence between VE and Vrnax is small and, In fact, is usually neglected. However, if a little more accuracy is desired, the equations for p and Vmax may be solved by successive approximation. The remainder of this section considers the motion oi the vehicle during the deceleration phase. During this

Preceding page blank

T"ME JC'1MI| NOIPINI

LJNIVEMS|I

APPLIEU PHYSICS LAB'RAIC)RY

period the general behavior of the body may be obtained by simpli[L-d equations of motion based on the following assumptions: 1. 2. The earth is flat and nonrotating. The only force acting on the body is the axial orcu, and the axial force coefficient is constnnt. The reentry flight path angle is constant.

:3.

The reentry phase is of relatively short duration and short range so that the assumption of a flat nonrotating earth is justified for this study. The assumpti.on that the axial force is the only force acting on the body implies that the angle of attack is always zero and that the effect of gravity is negligible compared to the effect of axial force. More precise methods of computing trajectories may be used to show that the effect of y Is usually small, Except for the very-high-altitude and, sometimes, the very-low-altitude portions of the reentry trajectory, the assumption that

W sin V
A

is small is justified for approxi-

mate solutions to the equations of motion, Precise methods of computing trajectories may be used to show that assumption 3 is valid for a large portion of most trajectories and that the validity tends to be improved for high weight and low axial force bodies and for steep reentry angle trajectories. For many reentry bodies and trajectories in current use, the assumption of constant V' is valid all the way to impact. The geometry of the trajectory in which A is the only force acting is shown in the sketch.

A)
S~~-~

1161

1-

I
IN& JOHNS HOPKINS U"Iywsl3m y

ApI'LIF:D PHYSICS LABORATORY

A HORIZONTAL

7
V

IS NEGATIVE

DURING REENTRY

hI

R.NGE

The equation of motion along the flight path is: -A ---'dv (4. 3-1)

Exprebsing the axial force in terms of coefficients as discussed in Section 2:

-CA (g 2)pV S

WdV

-T -

(4. 3-2)
/1 V sin ^

Since the flight path is a straight line, dh/dt and Eq. (4. 3-2) may be written in the form:

dV
V

CagPS dh
2W sin Y

(4. 3-3)

The term W/CAS occurs frequently in reentry equations and is called the ballistic coefficient, A. For an atmosphere in static equilibrium (see Eq. (3-1)), dP = -pgdh. When these substitutions are made, Eq. (4. 3-3) becomes:

IN

117-

.I

III$11 V THI JOHNS .HOpolNI UNSV4

APPLIED PY6ICS
mILVIN 59n140

LABORATORY
MAiPTLNNV

dV _ dP 203 sin Y V
Using initial confitions V Viat P

(4. :3-4)
Pi, Eq. (4. :3-4) Is

readily integrnted to give:

V2 V where K is defined by the equation:

(4. 3-5)

-P
KV
2

(4. 3-6)

Equation (4. 3-5) is valid provided

eE

- << K

This con-

listed above. dition is a restatement of assumption 2 invalid at low rebecome to tends Therefore the equation high values of K. entry velocities and at very low and very is not serious and The limitation at very low values of K section. On the was discussed at the beginning of this of K may have other hand, the limitation at high values 2 for example). 1. significant consequences (see Section 5. 0 and the equations When the initial altitude Is high, K.i of these equamay be simplified, For some applications altitude condition. tions the initial condition is not a high is retained in Therefore for completeness the K1 term Eq. (4. 3-5) and subsequent equations. may Other trajectory characteristics of interest Eqs. in given are be derived (Ref. 4-1) and the results from altitude hi (de(4, 3-7)through (4. 3-29). The time by P) is given fined by Pi) to any other altitude (defined by: ave
I

t t.

_ In -+
I

I (K)

K(4.

3-7)

S118

I
?iii

JOHNS HOPKINS UNIVeA"61 SILVIM SPRINl1 MA&T.ANV

APPLIED PHYSICS LABORATORY

where the I1 function is plotted in Figs. 4-7 and 4-8. dynamic pressure is given by the equation:

The

q:2 C AS11

2 C

si

eK - Ki

(4. 3-8)

The longitudinal acceleration is given by:

KV

sin Y
c K-KL

gx

(4. 3-9)

The Reynolds number based on body length is given by: ViL ve (K/2-Ki-/1 (4,3-10)

The heating rate at the body stagnation point is indicative of the severity of the heating environment. This rate is given by:

412O)j

e 3 (Ki

K)0

(4,3-11)

The total heat absorbed at the stagnation point from h1 to h is:

-1.69 Q =.VE (y6 000

sn ~*

12 (K

12 (K)

(4

-2

where I 2 (K) is plotted in Fig. 4-9. The dynamic pressure, longitudinal deceleration, Reynolds number, and stagnation heating rate all have a

119

JOHNS NOI-KINII uilVEIlrtY MTI APPLIED PHYSICS LABORATORY *1L, VlN SPRING MA NI1.1&nI

trend which increases from zero at reentry to a maximum value and then (depending upon the body and trajectory characteristics) decreases. The approximate pressure at which the maximum values of these functions are reached mny be obtained by assuming T is n constant, KL = 0, seLting the derivative of the f'unction with respect to K to zero, and solving foe' the maximum, The error introduced by assuming T is 0, VI = VE and the equations a constant is small. At K1 for the maxima and their X, P, and V/VE values are shown below, Equation

K
1

P
-0l sinY

F.
0.80Oi (4,.3-13)

S"
q

-fVE 2 Sip,/ 9330T

2 V i

s~an Y

0,606

(4,3-143)

gX"9330T

!-R

I,.

0', 8

2 nn

0.8

43-5

/ 8 o20(0/3)

T-16) 1/0

sin'

0,646

(4,

The longitudinal acceleration is a characteristic of considerable importance since it is a quantity that can readily be measured on a flight test vehicle. It is interm esting to note that for an isothermal atmosphere is highly dependent upon VE, moderately dependent upon y and T, and independent of the vehicle weight and axial force (p). However the altitude at which The greater the 0 or the. occurs is dependent upon , lower the altitude for the steeper the reentry angle, gXmax'

'1
Ii,

-120-

'C

"'a65 JOHNI$

"IOPkINI UNIV|II1if1?

APPLIED PHYSICS LABORATORY

The gx versus time curve has two inflection points,


which may-be determined in the following manner: d g X dt

(4. 3-17)

It mny bu shown that: dK T " E sin h K Ke -K/2

.-

(4.3-18)

!ind
-

E ..
2g(H)2

K(s

K)e 3K/2

(4, 3-19)

The inflection point occurs where Eq. (4. 3-19) is maximum

Sg9x

or minimum. and K - 2,

It may be shown that this occurs at K = 1/3 0. 650 and 0. 736,

which correspond to-gXmax

respectively, or P = -(1/3 )g sin / and -2g sin Y, respectively. Using Eq. (4. 3-7), the time between the high altitude inflection point and gXmax is given by: 80T

At 1

ave sin Y

(4.3-20)

and the time from gXmax to the low altitude inflection point is given by: 77T ave V sinY

t2

'

(4.3-21)

121

-_

_I ---

---.-

TM9 JOHN$
Sli..* Vi

NOPRING UW IrMITU,
*, lN6 M&AALAND

APPLIED PHYiICS LABOIATOIRY

From Eqs. (4. 3:-9) and(4. S'

14)(

2 (4. 3-22)

g --

gx

' 2. 72

Therefore, using just the telemetry gx trace, the ratio of gX/gXrnax is a good indication of the K history. The K history can be converted to air pressure history from known p and 7 (Eq, (4, 3-6)), and the pressure history can be converted to altitude history from known atmnospheric characteristics. A plot of gX/gXMax versus K is given in Fig. 4-10. Other quantities required for succeeding sections are listed below for convenient reference,

2
WA

SVE 2 CNCe AXm sin (-7


2 CAIH

K
eK (4,3-23)

1/2
kA 2

tir

A E
D

CAIYI a H

]
]k1/T2

(1

(4.3-24)

(4.3-25)

Ix

ly (4.3-26) :

=C

- 122-

__

THU[ JOHNP4 MOPkI~li UN9VURUIIT'U

4FAPPLIED PHYSICS LABORATORY


*IYU Sel mai.il M&*,T~AO

q-

2 V 3m sin7

(1 -K)
e

'K T2/

(4.3-27)

2CAS H

V2
K

E 2 2CABH rn (-,y)

(0 eK - K)

(4. (, 3-2B) B Sd

A
V
2 WVsinn y

CASH
A

77143
e K(4.3-29) E

The 1962 Standard Atmosphere characteristics required for evaluating these equations are listed in Table 3-1.

- 123-

I-.

Twe JONI*$ MOPNING UNIVtmIITV

4APPLIED PHYSICS LAIOCIATORY


SILVIO
SPR.IN MARYLAND

f0.J

.;0

0.7

11(K

IK 2

0.6

n-I 0.2*1

n-ni
'--

0.1

0.2

o.
04

0.6

0.8

1.0

1.2

.L
1.4

Fig 4-7

VARIATION OF i. (K) WITH K

Preceding page blank

125

7143 JOHN$ MOPIMINS UNIVIRSI,7

APPLIED PI4YBICtG LABOR~ATORY


BILVUS OPOING

MARY6ANG

12

IN
1261

5A

"APPLIED PHYSICS LABORATORY


SILVIA NPinING WAV6ANfl

lH4

JOHNI HOPKINS

.JNIVIONIIY

1,0

...

0.9

0.7 0,B

0.4

0.3

FIg.4.9

VARIATION OF 12 (K) WITH K

-127-

___

*
*

~APPLIED

Iota JCMINS I4oloiNg UNIV90191?V

PH'YSICS I.ABORATORY
MAIMANO

*~tV3~SPOI140

I...

LI.

m*

C1

0O

0e

-128-

I'4E JONBNll1OUINIl UNI!(g~bIlTV

APPLIED PHYIICU LABORATO0Y


MILVnlmD ING. MARTLAND

I. I.

Is.

4.4 EFFECTS OF BODY DYNAMICS

:}

12
I~
129 -

I
?HS JOHN
SILViE

NOPKING UNWkl RSITT


1111114

APPLIED PHYSICS LABORATORY


MAMiLAND

SYMBOLS FOR SECTION 4.4

Symbol a C1,C2 C 3, C4 , C CA

Definition constant defined by Eq. (4. 4-52) constants in Eq. (4, 4-33) onstants defined by Eqs, (4.4-45), (4. 4-58), and (4, 4-61) axial force coefficient normal force coefficient

Typical Un its

CA
CN

C
N

ddCN/dF
N

C1
p Cm d FXi Fy, F F g gL . q

roll damping derivative


pitch damping derivative reference length feet pounds lb/rad ft/sec2

aerodynamitc forces along body axes X, Y, Z CN qS acceleration of gravity load factor = N/W scale height v RT/g or angular momentum

feet 2 slug-ft /8sec 2

I0

PIx YIy J K k1 k2

moments of inertia about principal

axes X, Y, Z
zero order Bessel function -P/(A sin>') Y); constant defined by Eq. (4.4-30) constant defined by Eq. (4. 4-31)

slug-ft

rad/sec: rad/sec

1131

Preceding page blank

?K1 JOHNS6 110kiNG UWNVjIR6TV

APPLIED PHYSICS LABORAIORY


*,Lv~a 5.*.,O MAI., *NI "

o'-w-

Symbol 1, ni,1n

Definlition

Typical Units

aerodynrinniic rolling moment, pitching moment, nid yawing moment, respectivelv mnss - W/g roll damping factor, C a

lb-ft slugs lb-ft-sec

m M

'1
'4

I2

V -2

pitch damping factor, C

qSd2 V

lb-ft- se

2
lvi Mr yaw damping factor, C Vnr 2 + F lb-ft-sec

N P p, qj r q R S T t u, v, w

total lateral force, 4 ambient air pressure

pounds lb/ft2 rad/sec lb/ft 2

roll rato, pitch rate, yaw rate, respectively dynamic pressure gas constant = 1716 reference area temperature tirn.e missile velocity along X, Y, Z axes, respectively

2/sec
ft 2 GR seconds ft/ see

V
W X AX Xg.

missile velocity
missile weight constant defined in Eq. (4. 4-40) X c. p. - X e.g. distance from missile nose to the c. g.

ft/ see
pounds

feet feet

F'A2 C. .

Ilk.r

Ii
tZe JOMN 4OPIMW UNIVINGIIlv APPLIED PHYSICS LABORATORY

Symbol X ry c.p.

Definition distance from missile nose to the c, p, angle of ;tt, ck in the XZ plane

Typical Units feet radians

vector angle of attack

_ne2 +

radians
radians lb/ft 2 degrees degrees degrees degrees rad/sec rad/sec rad/sec rad/sec rad/sec seconds seconds degrees

angle of attack in the XY plane or ballistic coefficient, W/C AS flight path angle, negative for reentry trajectories ( angle defined by Eq. (4. 4-22) defined by Eq. (4. 4-23) angle between the missile X-axis and the local horizontal g WT W A frequency defined by Eq. (4. 4-21) transverse rate =Vq + r

0angle

2 +qr + total spin vector = aerodynamic frequency defined by Eq. (4.4-36) defined by Eq. (4.4-49) period defined by Eq. (4. 4-24) period defined by Eq. (4. 4-25) angle

Sfrequercy
"B T

Sphase
Subscripts E o

refers to conditions at entry refers to conditions at t = 0

A dot over a symbol means the first derivative with respect to time. Two dots over a symbol mean the second derivative with respect to time.

133-

T1I4 JOHN$ MOPKINB UINIVIrIINITV

APPLIED PHYSICS LABORATORY

SUMMARY AND CONCLUSIONS

In the preceding suction, the effects of the atmosphere on vehicle velocity and related factors were cons idered. In the present section, ittention is given to the effects of the atmosphere on the dynamic behavior of the vehicle, We are concerned here with the rotational motion of the vehicle about its center of gravity over relatively short periods of time. Six-degree-of-freedom equations of motion are presented and interpreted in terms of body motion characteristics in the absence of aerodynamic forces (for example, motion in the exoatmosphere, where gyroscopic motion is importani) at low altitudes where aerodynamic forces dictate the motion and at high altitudes where aerodynamic and gyroscopic forces are both important. Dynamic motion is important in the analysis of body heating and lateral loads. Equations are given for body rotational frequencies that are observed on telemetry traces from on-board equipment or from tracking radar signal strength traces. Numerical illustrations are given in Examples 3 and 5, Section 4. fJ. The major conclusions are: 1. In the absence of aerodynamic forces, a reentry-type body that is spun about its longitudinal axis will follow a motion pattern determined by conditions at the initiation of the spin. If spin is applied so that neither pitch nor yaw rate is introduced, the vehicle will maintain a fixed attitude in space. If a pitch and/or a yaw rate (sometimes the vector sum is called the transverse rate) is experienced when spin is introduced, tUe missile's longitudinal axis traces a cone whose apex is at the missile center, of gravity. The magnitude of the corne angle and the period of motion are functions of the transverse rate, missile roll rate, and moments of iertia about the longitudinal and transverse axes.

Preceding page blank

- 1 135 -

THE JOHN*

iOPtOINI UNIVaIFITY

APPLIED PHYSICS LABORATOHY

2.

If the body contains rate gyros, the body rate

observed on the telemetry trace during the exoatmo-

spheric portion of flight will show a sinusoidal trace,


The half-amplitude is the transverse rate existing at the time of spin-up, and the period is a function of the roll rate and the moments of Inertia about the longitudinal and transverse ,ixes.

3.

During the low altitude portion of the reentry

phase, aerodynamic forces dictate the motion of the vehicle. Generally, a missile reenters thr, atmosphere with some initial angle of attack. To prevent excessive lateral loads and heating problems the missile is designed so that the aerodynamic forces cause the angle of attack to approach zero rapidly. It is shown that for a symmetrical missile the angle of attack follows a damped oscillatory motion provided certain static and dynamic stability criteria are met. The frequency of the oscillation is called the aerodynamic frequency. This frequency increases from zero at reentry, reaches a maximum, and then (depending upon the trajectory and reentry conditions) decreases until impact. 4. At high altitudes, both aerodynamic and gyroscopic moments are important. The gyroscopic moments tend to maintain the body at a fixed altitude; the aerodynamic moments tend to rotate the body so that zero angle of attack is maintained. Typically, gyroscopic motion predominates at altitudes above about 300 000 feet; aerodynamic moments predominate at altitudes below about 100 000 feet; both effects are important at the intermediate altitudes,

-136-

?Ir

JOlNP4

MOPKINS UN.V3Nt9 ?Y

APPLIED PHYSICS L.ABORATORY


ISILVES O3iNING MARVLANM

In Section 4, 3 the body was constrained to motion


along a straight-line flight path. Therefore the equations of motion were reduced to a one-degree-of-freedom problem. In this section, we consider the six-degree-of-freeldom problem. However, the equations are simplified by making the following assumptions: 1. 2. The effect of gravity will be neglected. The body is assumed to have symmetry about the longitudinal axis.

The primary interest in this section is to consider how the body behaves over a relatively short time span. For this purpose the effect of gravity may be shown to be very small. By assuming the body to be symmetric about the longitudinal axis, the principal axes coincide with the body axes, and the equations of motion are simplified since the product-of-inertia terms are zero, The equations of motion are presented in a body axis system (Fig, 4-11) which remains fixed with respect to the body. The body rate terms In the equations of motion are the rates sensed by I-ate gyros mounted in a vehicle; the body translational acceleration terms are those that would be measured by accelerometers mounted at the vehicle center of gravity, The six equations defining the six degrees of freedom are: NX = I - qr(Iz - I 4I = m -pr(Ix

(4. 4-1) (4,4-2) (4.4-3)

- I)
1X

'IZ = n -pq(Iy

7,
F S 9

y
+ vr-wq

X
(4.4-4)

-137-

I,

THE JOHN~tSHOMPING UNIVKNURITY


APPLIED PHYSICS LASORATORY
IILVlA RiPM*INQMARVLAND

= Fr g + +wp

-ur

(4.4-5)

Fzg
+uq - vp
,

(4.4--6)

The auxiliary equations required are: I =M p


mn = F Z AX+ M qq

(4. 4-7)
(4.4-8)

n a -FyAX + M r y
u 2 2

(4.4-9)

V:U

+V

+w

(4, 4-10)

_ Vv2 +w2
tan; FX = -CA'S Fy -- -F 2"v -F,( (4.4-11) (4.4-12) (4.4-13)

IV2 + w2

F7

-F

(4.4-14)

%V +w

4CLqS

9s

(4.4-15)

138 -

Zall

JOHNSMHOPKINS

UNIV90111"

APPLIED PHYSICS LABORATORY


SILVIO SPRONS4 MARILAND

4.4.1

Zero Aerodynamic Forces

Consider first the case of flight in the exontniosphere where nil nerodynamin terms are zero. In this region only gyroscopic effects exist. Since the body is assumed to be symmetrical, Iz - 1y and Eq. (4. 4-1) reduces to the condition where p equals a constant, Differentiation of Eq. (4. 4-2) gives:

,4

+ pP

Iy

(4,4-16)

Substituting for Iz = Iy p2+


+ p

f, from Eq. (4. 4-3) and noting that I )2 IY "

q = 0

(4.4-17)

The solution to this differential equation is:

COS CoT

T II

p(l - Iy) t +t

If time is measured from the time when q 0 and ; I qwcos p(

qmax

then

-)

(4.4-18)

Y
SW OP(I ---r)

sin

p(1

IX
-

-139-

-I
S'
" ' "I - ',,. ,- . .

' -- 1 . .

71Iq JOHNS HIOPKINS UNIVIrpSIty

APPLIED PHYSICS LABORATORY


5I1.4,111" fPool" MAMfLAND

Substituting into Eq, (4.4-2): IX r


.Cl

sin p(1 "2


+I r

t 2

(4. 4-1U) (.-0


4. -0

=W

Therefore, the pitch or yaw rate measured by a rate gyro mounted on a vehicle is a sLnumoidal oscillation of halfamplitude WT having a frequency:
IX p(l - Xg ,Y)

(4.4-21)

This equation is convenient for checking the calibration constant for the roll rate gyro in flight test work. The term WT is sometimes called the body transverse rate. The solution to the equations of motion in body axes provides information regarding the time history of quantities that may be sensed by "on-board' instrumentation, but no information is provided on the motion of the vehicle in space. The motion in space is obtained by solving the equations of motion written In the inertial axis system. It is shown (Ref. 4-2) that, in the absence of aerodynamic forces, the motion of the body may be described physically as a "body cone" rolling along a "space cone" (or "cone of precession"), as shown in the sketch, page 141.

-140-

All.
__"_____ __,___ ___"____

II
?T"e J)ING MOORPING iJNIVVINNITY

APPLI[

PHYSICS LABORATORY
SlLVIR SPRING MANVL.Ako

BODY CENTERLINE

PP
1 Y of the space cone is the angular The centerline W~ momentum vector which remains fixed in space. The centerline of the body cone is the vehiclets longitudinal axis of symmetry (roll rate axis). The semicone angle

of the body cone is defined by the equation:


Th etrieof the space contan cone is andgulgvenby

tan =

(4. 4-22)

tan ( where is defined by bT enq. (4.4-20)

(4, 4-23) It should be noted


+ + r,

that the magnitude ofthe total spin vector,b y:

S~-

141 -

'I

I
TVII 10MfphlMPIgM I I UIM L~i vttIq vlllT

APPLIED PHYSICS LABORATORY


SInVIN SPRING MAPLAND

is constant and lies along the space cone surface, Assuming a positive roll rate and IX < I , the body centerline rotates about 1-1in a clockwise dyrection (as shown above) when viewed by an observer looking toward +H. The transverse rate vector rotates about the body centerltne so that the period is given by: T (4.4-24)

and corresponds to the frequency given previously in tEq, (4.4-21), The body centerline processes about the space

cone with a period:


2l
...

(4,4-25)

Ty p V I + tan2

For reentry bodies of current design, rs is an order of magnitude greater than rB. The utility of the physical picture described above may be illustrutted by the following example, Suppose we wish to determine the coning motion for n specified set of conditions at reentry, In particular, select a condition at t -. 0 where 9 = ; and lies in the body XZ plane, which is the plane of the trajectory. Consider first the case of qo 0 -and r ro, where r is positive and may be represented by a vector along the +Z body axis. Since W lies on the surface on the space cone and the body cone is external to the space cone, the angular momentum vector lies in the trajectory plane and E varies from W. (a maximum value) to 20 (a minimum value). Similarly, if ro is negetive, b varies from o (a minimum value) to a maximum value of -o + 20, The body-fixed Y-axis at t - 0 is directed toward the right of the trajectory as viewed by an observer looking toward the impact point. Therefore for an initial condition of pitch rate only, if q is positiv,.
-o

142

?bH JOHNG MOOKINO UNIVIINSIT

APPLIED PHYSICS LABORATORY


GLlVlI OPNINS MI-tUANU

I}
p X BODY CENTERLINE

HORIZONTAL

-H

JV

I
the angular momentum vector is directed to the right of the trajectory so that the nose of the vehicle remains in or to the right of the trajectory plane, Similarly, if qo is negative, the vehicle nose is always in or to the left of the trajectory plane. 4. 4. 2 Zero Gyroscopic Forces Consider next the case where the aerodynami.c terms are much greater than the gyroscopic terms (body rate terms). In addition we will make the following assumptions: 1 The trajectory region of interest is over a very short time period so that the dynamic pressure is constant. 2. 3. 4. The roll rate is zero. The pitching motion is planar (I. e. , only pitch or only yaw motion),

ey is small so that the use of linearized derivatives is valid, and cos 0 u 1; sin .,

Suppose that at some point along the trajectory the vehicle i- perturbed to some a * So We wish to study the

-143-

'.1

tHS JOHK1NSM040 IN5 UNIViRSITY

APPLI-D PHYSICBS LABORATORY


SILVIA WSIN* MAVIILAND

resulting motion of the vehicle under the influence of just aerodynamic forces. The geometry of the trajectory Is shown in the diagram,

HORIZONTAL

0'

S"Yo

ALL ANGLES SHOWN ARE NEGATIVE

BODY CENTERLINE

From Eqq. (4. 4-2) and (4. 4-8):


INy MFAX + M qq .

From the geometry of the trajectory:

, .

(4. 4-26) 1,

From Eqs. (4. 4-6) and (4, 4-26), using a small a approximation, and noting that u may be shown that:
.

V cos

and w =V sin;,

it

.'

'WV

(4.4-27)

144-

1i
1T4K JOHONP MOPKINI0 UNIVtSIQI ,T

APPLIED PHYSICS LABORATORY

l Pq mP tIIPv MARV6IA NO

;Using the linear relationship between CN aRnd 7: Z CN s

PI
Z

CN qS
qS

(4.4-28)

FZ - -N

Using Eqs. (4. 4-25) through (4, 4-28), it may be shown that the body motion is defined by the differential equation: o'+ k 1o + k 2 ev where 0 (4, 4-29)

CN

Cs
wv g

kl "-*W 1i

C "--* -

C sd CM
q~d
(4, 4-30),.: I V(44-0

CN 6XS -a -

C q

CN

(cSd)2 (4.4-31)

YW g

V2 Y

For bodies of Interest, the C

term of Eq. (4. 4-31) Is mq very small compared to the CN term, and the equation for k2 becomes: k2 CN XqS
42 (4,4-32)

ly

145

!i

i*

TM41JOam"l MORNINUNlyNIJ VIlITy

APPLIEO PHYSICS LABORATORY


SLIN lPOINO MARYLAND

For statically stable vehicles, k 2 is positive, and the solution to Eq, (4, 4-29) is:

1C, sint

+ C2 cos t

2 "-4

(4. 4-33)

where C 1 and C 2 ate arbitrary constants deiined by the initial conditions. Let No and F - 0 at t = 0 so that C2 = ?o and C1 = 0 and Eq. (4. 4-33) becomes:

k2
cost

(4.4-:34)

The cefore, for a statically stable vehicle, the n history is oscillatory and is damped (converges to ,-oro) provided k1 > 0. if k1 < 0, ; diverges even though the vehicle is statically stable. Some flight tests of reentry vehicles have been observed to exhibit this type of ; divergence. To assure W convergence, then: k > 0 or AX 0 static stability2

(4. 4-35)
> 0 dynamic stability

k > 0 or C 1 N

Md
-

m qI

The frequency of the oscillation is given by the

equation:

WA

k2

"W

|k

"-146-

Tk-K JOHN$I -OP.'N9

o'lt

s'RII'I

APF'LIOED PHYSICS LABORATORY


Sh,f
*PR4Q0 UAmrLA.ND

k 'or most bodies 1. >> = and the fkrequency hecomes:

A X (IS
-(

4.. 4-.Ar,)

A
This frequency, called the aerodynamic pitch f'equency, is of gront importance in the motion of reentry bodies is will b. shown later. It should be nAted tha-t is Zero :d reentr (C, = 0)and, for constant CN and W AX, W A is proportionrl to lq.

A solution to the equations of motion (Ref. 4-3),ln which the d.)namic pressure was allowed to vory, shows only small d[ffermenes frorm the results given above cxcept for very blunt bodies and nearly zero values of X. The aerodynamic frequency may be evaluated as a function of altitude (T, K) using equations derived in Section 4. 3. From Eqs, (4.3.8) and (4.4. 36):

2 A

VE2CN 4Xm sin(-y) clK 2 CAIyIH (4.4-37)

4.4. 3 Nonzero Gyroscopic and Aerodynamic Forces At very high altitudes the dynamic motion of the body is governed by the gyroscopic terms in the equations of motion; at low altitudes where the dynamic pres sure is high, the aerodynamic forces predominate. We look next at the transition from one region to the other. Using the equations of Ref. 4-3 and the reentry characteristics given in Section 4.3, it may he shown that r 0 at reentry the ty history is defined by the for p = q equation: 147-

TM& JOHN*

HOPiKINS UNIVErRII.

APPLIED PiYSICS LABORATORY

ry ry L

ll

2 WA Vt:E= sin 'ye 2 -1

(4. 4.-3 B)
2C

CXPI-(C
J4( A

-C

-CI

q Y

where Jo(X) is the zero order Bessel function of thi first kind plotted in Fig. 4-3 2. Writing WA in terms of K using Eq. (4.4-37):

2Hm
-

N, CA1 Y " Cm C

1 , (4. 4-:39)

Jo E C c xp
(V

sin,
K [Tr

(C2 N

Iy

A
2C

qY n

Therefore, for a given vehicle and trajectory, the C! history Is a functLon of P and T only and P is the predominant factor. Even without the exponential damping function in the denominator ofEq. (4, 4-39), the ey history is a damped oscillation about y = 0, and the frequency is the aerodynamic frequency, WA. For conditions near entry (4- ;-- 1), K - 0 so that the exponential term in Eq. FeE (4. 4-390) 1, and writing K in terms of P:

I.

I
S2 01Eg
_

~2SF!
sin y Y /

Jo(X) .

(4, 4-40)

Equation (4. 4-40) may be used to determine the altitude (P) at which the atmosphere begins to affect the body motion. Arbitrsrily selecting vy/e.E = 0. 99 as the eondition when the atmosphere begins to affect the body motion, from the Bessel function tablesthe argument, X, corre0.99 is 0.2. sponding to F'l E 1448-

HIK JOCMNAMOPKINS UNIVCNSI1Y

APPLIED PIHYSICS LABORATO"RY


$i.16VR SPRIING MAMV'LAND

'' P

2V Iyg sin" 0.02 -9C


NX

4.4-41) (X

The altitude for the first peak in a corresponding to X = 3. 83 shown in Fig. 4-12 may be obtained from:

sin 2 ' 7. 34 1 g 2
Iy
SHCE AX N

(4.4-42)

The ; envelope could be obtained directly from tables; however, for ;/WE less than about 0. 8, Jo(X) Jo(X) is approximated quite closely by the equation:

J (X)

(X 0o

x)

(4.4-43)

Therefore, the envelope corresponding to Eq. (4. 4-38) is:

-V

sinv

e3

where

c3= 4\C A

Iy2 c A md
AA

(4.4-45)

and is primarily the aerodynamic damping term. K C3 K 1: altitudes where eK and e

At high

-149-

Tt41 JOHN$

"OPKIINI

UNIVIlSyT

APPLIED PHYSICS LABO9RATORY

2I

s In

~(4. 4-46)

The ry convergence at high altitudes is a function of body ), aerodynamic charmass ehnracteristics (I, S, X

acteristics (CNd Xc, p,,, type ofki'ajectory (YC) and ntinospheric properties (T, P). For a given trajectory, mass, and aerodynamic characteristics, a Is inversely propor-

tional to

1-'

or approximately toA P,

V or a given

mass, aerodynamic charactcristics, and altitude, the That is, at a given convergence is proportional to V altitude, the o for a shallow trajectory will have converged more than that for a steep trajectory. Increasing CNI and AX tends to incvease the convergence at a given altitude' increasing Iy/S tends to decrease the convergence. If p A 0 at reentry, the early ; history is changed, The gyroscopic terms tend to maintain the body motion about the angular momentum vector, which is fixed in space and usually is not directed along the velocity vector. The aerodynamic terms tend to cause oscillation about the velocity vector. As a result, in the Initial phase of reentry, i oscillates about an angle of attack which, itself, is gradually reduced as the aerodynamic forces increase. The equations defining and / for the case of WT = 0 are (Ref. 4-4):

S-

A cos

--

-4t (4.4-47)

S-A sin _wt

-150

I
UNIV911181 JOHNS HOPKINS APPLIED PHYSICS LABORATORY
SIl*1R
SPSNG

TiH

M&ftI.IMLA

X*~~-'

2
y

L-x
y

(4.4-49)

In the intttal phase of reentry where WA << Senvelope is given by:'

p, the

Owl-I

"A xP

II

(4.4-50)

iy
Where WA shown (Ref. 4-5) that the ratio a roll is given by the equation: is not small compared to

1
IX p, it may be

I1

21

itmy
without

with roll to

a tanh a &Poo
HpIx

(4.4-51)

"a

Eysin

(4.4-52)
1, and the 1

For values of a greater than about 3, tanh a a history corresponding to Eq. (4. 4-44) is:

151-

.IL
;,, : . :' .'l..,,.. /... i.- .

TH9 JOHNS HOPRINi UNIYvmIITV

APPLIED PHYSICS LABORATORY


SILMIN SPRING MANTYAND

Y A

(4, 4-53)

e 3
C 3K 1, the o. enve-

Therefore, at high atitudes where e

lope converges to a a 0 in proportion to

A -A7w
J,
4

For a nonrolling vehicle at reentry, a begins to decrease when the aerodynamic moment Is sufficiently using Eq. (4.4-41). hlgh,and this altitude can be estimated For a rolling vehicle, re begins to decrease when the aerodynamic moment is significant relative to the gyroscopic moment, This altitude may be estimated using Eq. (4. 4-50) and solving for
aA AL

0. 99, so that:

WA

0.1 I Iy

(4.4-54)

for WA Substituting eK '

using Eq. (4. 4-37) and assuming


0O.02Ix 2

=
RT

2
- p2
2

0.

2 AXSVE

(4. 4-55)

1
T

I C

Y N
A rapid convergence in a is desirable in order to prevent excessive lateral loads resulting from the 3E. The lateral load factor is given by: S gL " N =CN E Wqo(E ( (4.4-56)

1
,"

-152

UNIVtftIIY' ?r~I ,JONNImMOPKI(INS ArrLIED PHYSICS LABORATOgRY


iILVlE 4PRII, MIA*,LA%

For a nonrolling vehicle, using envelope of 9L is Igiven by!

/Efrom

Eq. (4. 4-44)

and i from Eq. (4. 3-8), it may be sgown that the maximum

23
9L E 12_'

1/4 K1 Kl elI Ri CA rd

gH

c+
A

0.75
A

(4. 4-58)
IV

Neglecting the small variation In H with K, mUnn occurs at:

the maxi-

i4

K -3 :.

(4.4-59)

74

Using Eq. (4. 4-59) and substituting for the constants in Eq, (4, 4-57), the magnitude of (gL is given by:

maxI
2 0O,1 2 8"s. max
n Y)

C C-

3/4

1/4 (4, 4-60)

where C 52

CA + LC
A

N a4

2 m~
-a:

(A

C4

(4.4-61)

I
?HII JOHN|E HOPPINSl UNIVURI,,IT

APPLIED PHYSICS LABORATORY


SilkVift

S#~n*I

MIPIYI~k b

I
i

In evaluating Eq. (4.4-60), a- must be in radians. E The corresponding equations for a rolling vohicle at reentry are,
2 2 31/ 4

I
4-62)

IL

0 42

K(4,

and

3E) 114

3 /4

N(

3/4

1/2

w0.16

(4,4-63)

LTI;I
The maximum vnlue of gLfor a rolling vehicle also occurs 3 1 at Ku- . 4 The ratio of maximum gL for a rolling vehicle to maximum gL for a nonrolling vehicle, obtainied from Eqs.
I

(4.4-60) n-(4, 4-63 3

g Lp= 0

at' K
/ V[il l - 154-

L, 1. "2

H" X 1"

'

APPLIED PHYS'ICS LABORATORY


GILVIN IIIm,... MAOVLAP.P

U, V.

w VELOCITIES

A, Y, N F!ORCES N Im, n MOMENTS

NOTES:AXE SOFWNTC PSTVVANLUES a.


* wu 2. tana

mi41

BONOXS YTEM:

I OI EVLE HW
2.__tona_

You

t1 6 J COH N$ H 4O P K INS U dIVV IS 3IT T APPLIED PH4YSICS LAMORAORnY

U-l

IsI
I
~

15

?weU

~4044

UNIVISOITY wmoe~~

It

APPLIKO 10HYGIC LABORATORY

4.5 EFFECTS OF ASYMMETRIES

'157

I
rmi JOHN* PIOPKIND UNIVeypITY 61'VKW $04,000 MAftVLAND

APPLILD PHYSICS LABORATORY

SYMBOLS FOR SECTION 4,!5

Symbol A

Definition amplification factor in angle oi" attack

Typical Units

a, a1l a2
CA C C C C

constants defined by Eqs. (4. r.-B2),


(4. 5-87), and (4. 5-BB) axial force coefficient roll moment coefficient, 1/qSd roll damping coefficient, 6C I/b(pd/V) pitching moment coefficient, m/qSd
-

pitching moment coefficient at a --. 0

0
Cm CN Cn

i0

(45100a/e
bCm /(qd/V)

pitchdamping coefficient,

normal force coefficient yawing moment coefficient at

w0

N
C C 1 ,C c. g. c. p. D d F F

N
2

dC

/dy rad/sec

constants in Eq. (4.5-110) center of gravity center' of pressure damping function defined by Eq. (4.5-20) reference diameter 4F2 2

feet pud

pounds A
asymmetry factor defined by

Eq. (4.5-72)

Preceding page blank

159

. ....

...

.. ....

. ...

XP PLIED PHYSICS LABORATORY


SILVIN SPRING M&MVILAND

?MR JOMNII WPKIN&

I.NIVURPIYI

S ymbol FXPF F
&

Definition forces along the X, Y, Z body axes C qS N function defined by Eq. (4, 5-47)and (4, 5-73) factors defined by Eq. (4. 5-125)roll, amplification factor defined by Eq. (4, 5-37) G required for roll "lock-in"

Typical Units pounds lb/raid

F(X. D)

VV V2
F VF G GR GI

factors defined by Eqs. (4. 5-62)

roll amplification factor defined by Eq. (4. 5-80) of gravity gacceleration gXlongitudinal G1 G 2 11 11altiLude IA 4 A Iz moments of inertia about X, Y, Z axes acceleration

ft/ sec 2 g

roll amplification factors def'ined by Elqs. (4. 5 -81) and (4. 5 -84) scale height, lRT/g f eet feet slug-ft2

1'
K

Y, Z2 1 products of inertia about 1 5-101)) (Eqs. (4. 5-99) to (4, X, YZaxes xyX~z
-P/($ sin 7)

slug-ft2

k, k 1, in, n
0

constants defined by Eqs. (4, 5-59) and (4. 5-60) aerodynamic moments about X, Y, Z asymmetric moments abouit X, Y, Z Mach number roll damping moment/p

lb-ft lb-ft lb-ft-sec

I M In
0

M M p

-160-

AP,'L.IC

P0ipYSICS LASCRATORY

Symbol
M M

efinition
pitch (ha iutnF monent/q yiiw (1ch1pinp, vehicle Ln:SS mrioment/r

Typical
lb-ft-sec lb-ft-sec

UnatN

arnbient Mir pressure

lb/ft2

S~Probtibil~ty

n, q,
nj

body rates ,bout the X, Y, Z ax"s reduced roll rate


nrofl ,cceleratilon Eq, (4,5-3B)

rdAlsec / SCc

P'rad
term defined by

X'sd r8d)sec' 11..


2

SVe",

Ap

P " P(2)PWNI2 "dynamic pressure

gr, js constaint r
S T t V W

1716

f' I2 t 2 'i -Pcow s c pout...] j feet reet fe t feet rad ,ns radians

radlal c.g, offset, always positive


reference area amibient air temperature time vhictle velocity vehicle weight nose to the c.g. distance from vehicle nose to the c, p. distance from vehicle c.. c. p. ofaxis gravity offaet along center Z y Y nd angle of attack In the XZ plane angle of attack resulting from 0 in the XZ p].ane at X-

x
AX Yc. g, co, g.

16,..

*1O ,AA

TMIN JOHN$ HOPKIUNB

AP LIED PHYSICS LASORACRY

S.mbol a
_ p

Definitton total vector angle of attack +2


t2 I2 o)f ittrack in the XY plane
2

Typical Units

radians
radians dnn lb/ft2

~~anglu,

ballistic v-oeffic ient, W/CAS


8

or

angle of attack resulting from 8 XY plane at in the o -u co radians radians radians radians degrees degrees

ry
6inclinatinn

inclination of principal axis in the XZ plane


of pvincipal axis in the

XY plane 6 62 2

632

ang e defined by Eq. (4. 5-83) y flight path angle, always reentry negativedees or ratio of specific heat, 1. 4 space cone half-angle or angle defined by Eq. (4. 5-31)
sin 1

degrees degrees

0 C. g.

degrees

oM
Pr/WA
A cos A.Cos ] r/d - r/d)min

degrees

OI
-162-

(.<

THE JOHN$ HOP1INS UNIVUK4IYT

APPLIED PHYSICS LABORATORY


5LI.VR
UPI~lg

MAU4"..APSO]

Symbol

Definition
angle defined in Section 4. 5. 1 angle defined in Section 4, 5. 1

TypLca! Units
degrees degrees seconds rad/sec rad/sec rad/se:

period of oscillation aerodynamic pitch frequency

WA

Sp(1 - I /
*T Subscripts c. g. E e i M max amin o R S. L.

transverse rate vector, ' +

refers to c. g. or is caused by c. g. offset refers to reentry condition refers to equilibrium condition refers to conditions at t = 0 or to an elementary quantity caused by asymmetric moments refers to maximum value refers to minimum value refers to conditions at p = 0 refers to resonance conditions refers to sea level conditions

A dot over a symbol means the first derivative with respect to time. Two dots over a symbol means the second derivative with respect to time. An arrow over a symbol means a vector quantity.

tN

-163-

"41k

I
T"94 JOHNS MHOPKIN9UNIVS0ITV

APPLIED PHYSICS LABORATORY


mtLvYRN SPRING M.LART60

I
SUMMARY AND CONCLUSIONS A reentry body is normally spun about Its longitudinal axis to provide stability during the exoatmospheric phase of flight and to reducc dispersion resulting from phenomena that occur during reentry. Ideally, the roll rate remains nearly constant during reentry. However, under the influence of mass and aerodynamic asymmetries, the roll rate may change radically. Under certain conditions the roll rate may change so that the roll rate and the aerodynamic pitch frequency are nearly the same for long intervals of time. When this occurs the roll rate is said to "lock-in" with the aerodynamic frequency. This situation is sometimes disastrous since the angles of attack and thus the lateral loads become very large, The loading situation may be aggravated further by aerothermoelastic interaction. On the other hand, the asymmetries may cause the magnitude of the roll rate to decrease and change sign. In this case the missile will be driven off course in proportion to and in the direction of the lateral loads existing at the timne the roll rate is nearly zero. Even if the roll rate passes rapidly through zero roll rate, the impact dispersion may be excessive for high performance vehicles. The six-degree-of-freedom equations are modified to include various types of asymmetry, and the conditions under which excessive loads and/or dispersion may exist are discussed. The major conclusions are: 1. The effects on body motion of aerodynamic asymmetries (abnormal pitching, yawing, and rolling moments, and center of pressure offset from the longitudinal axis of the missile) and mass asymmetries (center of gravity offset from the longitudinal axis of the missile, products of inertia, and unequal pitch and yaw moments of inertia) are considered, All asymmetries considered

Preceding page blank


_q

- 165

,TH

JOHNG HOPKINS UNIV I

'IS lT

APPLIED PHYSICS LAeORATORY

have some effect on body motion. However, for reasonable magnitudes of asymmetry, the effects of unequnl pitch and yaw moments of inertia are negligible, the effects of any one of the other asymmetries are small, but the effects on vehicle performance of some combinations of the asymmetries may be catastrophic. 2. Unacceptable vehicle ;erfcrmance is likely If either resonance (resulting In large lateral loadi) or zero roll rate (resulting in large impact dispersion) occurs during reentry. In either case the asymmetries must cause both a trim angle of attack and a roll torque, Therefore, a center of gravity offset combined with either an asymmetric pitching (or yawing) moment or with nonzero products of inertia or a pure rolling moment asymmetry combined with either of these asymmetries is apt to result in unacceptable performance. 3. The conditions under which adverse performance is likely tv occur are dependent upon many variables so that each vehicle must be analyzed separately. Nevertheless, the following trends may be stated. The probability of encountering difficulties resulting from combined asymmetries is increased for missiles-, o o 'o o o Having low stability margin Having resonance at an altitude near maximum dynamic pressure Having low normal force per degree angle of attack arid/or a low pitch damping moment Flying shallow trajectories Having slender shapes

o
4.

Having small dimensions.


Detailed equations are presented for evaluating

the probability of encountering "lock-in" or spin through


zero roll rate for a missile of any arbitrary design. Comments are made regarding the accuracy of these equations. Illustrations of the use of these equations are given in Exnmples 5, 6, and 7, Section 4. 6. -166-

""

J"

Y141 JOHN$P OPKINS

UNIVKRGI

APPLIED PHYSICS LABORATORY


SILVER SPRlIIN MARI.*bND

4-

In the derivation of the equations given in Section 4. 4, it was assumed that the body was symmetrical. We consider next the effects on vehicle motion of asymmetry in mass and external shape. Asymmetries in mass and external shape result from design and manufacturing tolerances or from inflight phenomena and may cause one or more of the follow-

Ing anomalies:4
Aerodynamics
1. 2. 3. A change in normal force characteristics A change In pitching moment characteristics A lateral shift in center of pressure A pure roll torque. Mass 5. 6. 7. A lateral shift in center of gravity Nonzero products of inertia Unequal pitch and yaw moments of inertia.

4.

The effect of asymmetries on normal force is

usually negligible but the effect on the pitching moment


is very significant. A lateral shift in the center of pres'-ure has the same effect as a lateral shift in the center of gravity, and these two effects will be combined in the following discussion. For convenience, the effect will be referred to as a center of gravity offset effect, but it is important to remember that the same effect results from a center of pressure offset. This section will be limited to the following asymmetries: 1. 2. 3. Asymmetric moment, mo and n 0 Center of gravity offset, Z and Y

i .

C.g.

e. g.

Nonzero products of inertia,

and

L
" A,.
= . .

-167

. . .: i : : : , :? :: . , I

. .

..

T~~~YM JOiiNs MOPWIINB UNIV5NIIIIT APPLIED PHYSICS LANORATORY


liLY1111 ltlNS M&4y6,ND

4. 5.

Unequal pitch and yaw moments of inertia Pure roll torque, I.. A

The asymmetrLes have negligible effect on all portions of the trajectory except for the reentry phase. To account for the asymmetries, the equations given in Section 4. 4 must be modified to the following (see the sketch on page 170):
0b1X 0 1 - qr(IZ - 1y) +Xy(4l - pr)l+lxzlf + pq +l+Izlq" - r2 B4,-1)

i!

-prox - IZ) + -Xy(O + rq) + IXZ


t 1z Wn

2 -

I(+I

. pq)

(4, 5-2)
(4. 5-3) (4.5-4) (4.5-5) (4.5-8)
.1

lI) -

2 + 1y(p

q2) + 1 +1

-(0

- rq) + iyz( + pr)

P'F ,M Z -F Y Z cg P Y e~. Z q X 0.9.

"m"F AX+Mq-F Z
n

+tn

-FyAX + Mrr + FY

rc. g.

+n

4.5.1

Effect of Asymmetric Moment moment (mi


0'

that only an asymmetric and no) Assuming exists, Eq. (4,.5-1) reduces to: X = Mpp,

(4. 5-7)

Therefore the roll rate decreases slightly under the influence of viscous damping, M . The trim equations F+ -F qq+m AX + Mr +no =x+ pr(Iy-IX) r X (4.5-8) (459)

.,

r
-168-

Y
,I

AI

5
I

?Mt JawksI HOOKIGi

UNIVtNSITY RYLAND

AP.IED PHYGICS LABORATORY


4IlViN APIM tNS MA

Consider first the trim conditions for p * 0. The pitch damping terms are negligible for this case, and the trim

equations become: m
z
0

5-10)

AX n0 a0

(4. 5-11)
in terms of a and j (Eqs. (4,4-13)

Writing F avid F1 to (4.,4-15))and


C,

expressing m, and no ncefietor

(m . MCmoqSd and no

CnqSd).

o
0 Nd AX Cno eN

.(4.

5-12)

The total angle of attack is,,

[(C)2

+e(Cn)'I
" AX

_C___

cN
and the plane of
by the equation:

-N

t-f

AX

(4.5-14)

defined by the direction of Fo is given

cot
,-

Fy
0 o

Cn

"o -'0 169-F

(4.5-15)

',-.I

.-.-

,....

...

""

-,

..

..

...

....-..

,;,,.

1,149 JOHNn
APPLIKI

HOPKINS UNIVVG

?ITY

PHYSICS LABORATORY

V
I~I-

+Za'.F~
Fo

Note that for

o "0 011to

900

i3 s

-and as s

a 9 00 to 18O 00 - 1B00 to 2700 - 2700 to 3600

is + ands rysis + andp:I +1 i s - and ry Is +

14

Sattack,

The angle 0o also defines the plane of the asym,-!I metry. _That is, for zero roll rate the total angle of jyol is in a plane containing the asymmetry. For example, if the asymmetry results from a flat spot on Sa 0 or 0-00o + is 180 (depending 0 the body, flat located upon the fore and at a circumferential location aftis Location ofthe the asymmetry).

'' ;' Ii'

di

tpn

te,

The effect of ro roll rates to modfy the magnitude and to cause the total lateral force to be rotated out of the plane of the asymmetry. These effects are given by the followdng equations (Ref. 4-6) based on theory given tn Refs. 4-o and 4-8. 4 of

-170-

4F?MRJO14N| HOPKINS UNIIRItTy &PLIIEDPHYSICS LABORATORY


GILIN.1 SP~RING MARYLAIND

.1
.
2

I
+ D2x2

-1/2

2a

,I
0

1(1

(4., 5-16)

-DX A in -D, 2+ 22 (I (1 X +D X (4.5-19)

FV

) " Mly q': /


-1 IX

(4.5-20)

1-Y_

Pr
A

(4.5-21)

From Eq. (4. 5-1G) it is apparent that the aniplification 1'ictor increases from 1 at X = 0 to a maximurn value
[D (

221-1,2

at X

and then decreases to 0

at X- S. Thus, at X greater than about 'V2(depending upon the value of D), the trim angle of attack Is attenuated by the effects of roll rate. Typical values of D are of the order of 0 to 0. 2 so that Amax occurs near X w I and lID. The condition of = 1Is called the pitchAmax yaw-roll, or simply the roll resonance condition. Since 171-

iMT

O JCM ,J M

UN W IVIIVT PMNSwp~

APPLIED PHYSIC8 LABORATORY

0. 1, the resonance condition Is very close to the SIX/ly condition P/WA n 1. Note that near resonance the pitch dam~ping term, D, plays n very important roll in limiting 0.
The value of D is always positive for conditions (static and dynamic stability) for which IEqs. (4. 5-16) and (4. 5-17) are valid, Therefore, from Eq. (4, 5-17) for positive roll rates the plane of ;. for a rolling vehicle is rotated counterclockwise (A0 is negative) from 0o. At resonance, the plane is rotated 900 from oo at very high values of X, the plane is rotated 1800 froin 0o.

+-j

F
F

ISNEGATIVFE 134

zI
..

For example, assume that an angle of attack is produced by a flap at the aft end of the body as shown in the sketch on page 173.

I
'172-

1l

I<
T,K JMHNI HOPKINS

APPLIE"

UNIvntrpry PHYSICS LA8014ATORY

1 I
Y y .(

W-!(
Z

At zero roll rate, this flap produces a +4yor a forc:e F

along the -Z axis. As the roll rate increases in a positive direction, F rotates counterclockwise. At resonance F " FR t Fo/D and is directed along the -Y axis. As the roll rate increases further, F continues to rotate counterclockwLse and decreases in magnitude. In the limit (very high roll rates) F - 0 and is directed along the +Z axis. To further illustrate the behavior of Eqs. (4. 5-16) and (4. 5-17), Fig. 4-13 shows plots of A and 60 as functions of X, for D = 0 and 0,.2. Note that neither A fnor A0 changes very rapidly for X from 0 to about 0. 5. Practically all the change in A0 occurs from X I 0. 8 to 1. 4,
.

F0
.

For

'

greater than about 1. 4, 0

is attenuated.

The effect of damping is small except near X = I where the effect on both A and A@ is large. As the reentry body descends through the atmosphere, WA increases, and unless the body is given a very high spin rate the resonance condition will occur at least once during reentry. If o were zero, the resonance condition

-173I ,

fH't

JOHNGHOPMINSUJNIV9NNITY
1I6VIN SPRIP40 MARYL.AND

APPLIED PHYSICS LABORATORY

would have no effect on the trajectory. If the only risymmetry were the asymmetric moments (n,, and in ), the effect of resonance would normally be minor since p is cessentially constant (changes by viscous damping only) and WA would pass through the resonance condition too rapidly to pi'oduce significant changes in the trajectory. The a would inecrase somewhat over the ry value, resuiting in correspornding increases In body lift and drag, but the time duration would be short, Therefore, the effect of an asymmetric moment, by itself, has negligible effuct on the trajectory for reasonable values of mo andA no* 4.5~. 2 Offset Center of Gravity The effect on trim conditions of the center of gravity offset (again neglecthig M1) .'or p 0 is given by:

F .

Co

g,

(4.5-22)

Fy

Y X

c~.*(4.
AX

5-23)

Substituting for F,, Fy, and F7 from Eqs. (4. 4-12)


through (4. 4-1.5): C A z (4.5-24)

%C
C N

AX Y *A

(4. 5-25)

2112

174 -

INC JCOHIG WOOIKINS uNIVKPSII

APPLIED PHYSICS LABORATORY


5 SVIft iPpRNG
MARILANIO

Tho effect of roll rate Is to modify the magnitude


of o and to change the plane in which it acts in accordance with Eqs. (4. 5-16) and (4. 5-17). However, unlike the aerodynamic asymmetry, the center of gravity offset affects thu roll rate equation (sec Eq. (4. 5-4)). The effect on roll rate is illustrnted by the sketch.

IFATI

For zero roll rate, o, is in a plane containing the center of gravity, and F acts through the center uf pressure directed away from the center of gravity as shown by on". H-owever, for a positive roll rate the force vector F is rotated counterclockwise from F as explained previously, Therefore, F results in a negative roll torque that tends to decrease the roll rate, A similar argument for a negative roll rate shows that the center of gravity offset results in a positive roll torque, again tending to drive the roll rate toward zero, Therefore, the effect of the center of gravity offset alone behaves as a roll damping effect and, in fact, must be accounted for when attempting to extract C, from expvtrnmental roll rate decay data. For

p
-

175-

.......

.......................................................

"

lii

i:

i'i l

l i

ll

i I

....

Jil

' '

-.

THI JOHN$B HOPKINO wNIV NG-IY

APPLIED PIHYSICS LASORA7ORY


Si1v3R

Ullle'4d

S&q.AILA

reasonable values of the center of gravity offset, the effect on roll rate is small except near resonance where the roll moment arm. is maximum (see sketch above), 4. 5. 3 Combined Asymmetry

(Coand r)

Both the center of gravity offset and asymmetric momem' result in minor effects on the trajectory when considered separately. hlowever, this is not the case when the two types of awymmetry are combined. Using the described procedures, the trim angles of sttack at zero roll rate for this case are given by the equations:

Am ci
0

zA

S= - - :-..AX

(4. 5 -2: 7)

'

Cy
CA 0o d no
A

cN

(4. 5-B)

=:

+/ is given by the equation:


0

(4. 5-29)

The direction of F

V
cot 00
00

o -' " :.

. .. ..

- -

(4. 5-30)

M Ad
Z c. g. -76
-

The proper quadrant for 00 is again determined as indicated in Section 4.5,1. An Important variable for this combined asymmetry is the angle, 0, between the plane of the center of gravity and the plane of Y.

VMS JO.N06 MQM5*

JNIVlSSU ITV '

AMPLIED PHYSICS LABORATORY

I
F

-c.g. 0
where

(4.5-31)

otI"c.g. = 00 to

c. 9. g.. if Y is + and Z

(4. 5-32)

9100

is +

= 90* to

1800

if
if if

is

and z

is +

= lBO to 2700 = 270* to 3600

c.g. e.g. Y is -and Z is c.g. e.g. Y is + and Z is c.g. e.g.

It must be assumed that the center of gravity offset and moment asymmetries are completely independent variables so that 0) nay have any value. For purposes of illustrating the significance of 00 it is convenient to consider "in plane" asymmetries (00 = 00 or 1800) and "out of plane" asymmetries (00 900 or 270*). For the in plane asymmetry, if 00 = 1800. the effect is the same as that

-177-

.---

w :-

APPLIK9

PHYSICS LABORATORY

SILeR URiWOmlq MAPV&.IO

described previously for the center of gravity offset. If 00 = 01, the magnitude of the roll rate will increase indefinitely (5 is always positive) if the initial roll rate is positive and will always increase negatively if the initial roll rate is negative. This case is shown in the sketch.

Y.1

F (-pl

At p 0, Fo passes through the center of gravity. If the roll rate is initially positivrt F must be counterclockwise from F. as shown, so U... a positive roll torque develops driving p to a larger positive value. Since F must lie between and 00 - 1 800, the roll torque Is always positive. Similarly, if the initial roll rate is negative, the roll torque produced by the asymmetries is always negative. For out of plane asymmetries, the case of 0o = 900 is sketched for an initial positive roll rate. Let the asymmetric force be located initially at F (counter clockwise from F since p is positive). The roll torque is negative and p decreases. For decreasing p, X decreases (assuming wA is constant) and F rotates clockwise. The roll rate becomes zero when F - Fo and then the vehicle begins to spin up in the negative direction. This process continues

"-178 -

,I

'I

fl[c

JOMNiSm HopiIp

UNIVRII!y

API'LI ED PHYSICS LABORATORY

RI
Z

0F

until F rotatos

900 from F0 . This point is an eqailibrium roll rate position, since at higher negative roll rates the roll torque becomes positive, driving F to a less negative roll rate. This equilibrium roll position also corresponds to the resonance condition, For 002700 the same phenomenon occurs except that the roll torque causes spin-up to resonance at positive roll rates and therefore does not spin through zero roll rate.

The potential problems resulting from these combined asymmetries are: 1. *2. Excessive spin rates resulting in structural or payload failures Prolonged periods of roll resonance resulting in high lateral loads and high drag (structural failure and excessive impact dispersion) 3. Spin through zero roll rate resulting in large impact errors (see Section 5. 4. 2).
-179

h.!

:.."

APPLIED PHYSICS LABO3RATORY ?III UNI 14OPKINI -N. JOHNS


B,16VER SPAINIS MARVLAML,

-_

,II
Both In plane and out of plane components of the

asymmetry can have aisastrous results onvehicle performance. However, the in plane asymmetries are less likely to cause difficulty because sfgrnficant roll torques are developed only at conditions near resonance. Also, this type of asymmetry cannot produce the problem of spin through zero. 4. 5. 4 Effect of Comb.nod Asymmetries on Roll Rate All the potential problems resulting from a combination of asymmetric moment and center of gravity offT set listed in Section 4,5.3 exist because of changes in roll rate during reentry. In this section we will consider the roll behavior in more detail, If Iy - Iy and if the products of inertia are zero, the roll acceleration (14q, (4, 5.1))becomes: Mp+ Z -F+Y

Y 0g,

Z C. g.

(4.5-33)

Ix
Neglecting the roll damping and pure roll torque terms and expressing the numerator in terms of F, r, and 0: -Fr
'

sin0
IX

(4,.-34)

Since F

CN qS , Eq. (4. 5-34) may be written:

_qSo'r sin

0
IxI
.

~(4.5-35)

The roll acceleration may be written as a product of two terms: G-

(4. 5-36) Max -180-

Sil

?MI J-Nt

.O3PN1

UNIV94iITY

APPLIED PHYSICS LABORATORY

where
=G-

sin 0
0

-A sin 0

(4.5-37)

'I

CN qSU 0 ir Ce. - # max Ix

(4,5-3[3 )

The G term contains the effects of roll rate and 0o; the JnOmax term contains the body geometry, environmental, and basic asymmetry terms. Note that b maxis the positive roll acceleration that would exist for zero roll rate if the plane of a were normal to the center of gravity plane (that is, 0o - 270"). Since 0 = 00 + A0, Eq. (4,5-37)
may be written..

G = -A cos AO (sin 0 + cos 0 tan Ao)


0 0

(4. 5-39)

Substituting for tan A$i and A cos AOp from Eqs, (4. 5-17) and (4. 5-18), Eq. (4.5-39) becomes: DX cos 00- (1- \2) sin 0 G22-2o +D2 X2 (l -X) X 2 +D

(4.5-40)

0 as the roll equilibrium condition, For We will define this condition either Omax or G must be zero. For a 0 reentry body in the sensible atmosphere ( 0), is zero only if co or r is zero. Generally, neither ao nor r is zero and the roll rate of a reentry body having combined asymmetries will vary until G = 0. Defining Xe as the value of X at G = 0, from Eq. (4. 5-40), >e=e or Xe must satisfy the equation:

?HI6 aOHNSl HOPKINSI UNIVSNlSIl'y

APPLIED PHYSICS LABORATORY


6ILVIN 3041Ni MAIVILAND

tan 00

DAe DX 2
-

'

(4.5-41)

Equation (4. 5-41) may be solved for Xs:

-D cot 0

o 2
F

xe

(4.5-42) 00 or 1800): (4.5-43)

For very large values of D cot 00 (Oo xe .0,

For very small values of D cot 00 (00 Pw90* or 2700): e1 -

cot 00

or

-(I +

cot 0)0

(4.5-44)

Equation (4. 5-42) is plotted as a function of D cot 00 In Fig. 4-14. The roll equilibrium condition, or unstable; that is, if the roll rate is X changes slightly from Xe' If the roll at X tends to drive X back to Xe, the Xe, may be stable changed slightly, torque developed equilibrium condi-

tion is stable; if the roll torque developed tends to drive X. away from Xe, the condition is unstable. The Xe condLtion is stable if (dG/dX)X (dG/d%)>e is positive. given by:
-3~
2

It may be shown that dG/dX is

is negative and unstable if

'4
2 +A sini0[D, 1)24-X

D oo0 ( 2 -D 1

-182-

______________________

TMEi JOHNSI NOPSItNS UIIVgNIYII


-

APPLIED PHYSICS LABORATORY 5041006N MARII|AND

~~IlL.VII

Therefore, the condition for Xe stability is that:

F(Xe, D) > tan 00


where

(4. 5-46)

D[ 3X' - (2 - D2)X2 F ,( D -F 0., Q) =


-

Z
(4.5-47)

2 - x4 J 2X[(D 2 - 1)+ 2x

For in plane asymmetries the condition for stability reduces to: Cos

0oll

+(2-

D-)X 2 - 3

<

(4. 5-48)

Since Xe = 0, the condition is stable for 00 : 1800 and unstable for 00 = 00.

For out of plane asymmetries, the condition for stability reduces to:

sin 0X

(D

1)+ x2

o0

(4.5-49)

Since X. = 1, the condition is unstable for Xe = +1 and 0o = 90* or Xe. -l and 00 = 2700; the condition is stable u00, for Xe = +1 and Oo = 2700 or Xe = -1 and Oo = Another characteristic of interest is the maximum values of G. The term maximum is used to mean either a positive or negative maximum. Maximum values occur at values of X obtained by setting the numerator of Eq. (4. 5-45) to zero giving the expression: tan 0 = F(X, D)
,

(4. 5-50)

For 0Oo = 0 or 1 800, neglecting terms smaller than D2 , Eq. (4. 5-50) may be solved for values of X and Gmax giving the result:

-183-

, ?. ..

_.

.J

," " ..:-,,.v(-

TH4 JOHN| HOPKINN WNIVtIIVt

APPLIED PHYUICS LASORATORY


SI.VIR SPOIrNG MARYLAND

Xk

2 D ( N - -n-)

D!

(4.,5-51)

From Eq, (4. 5-40):


2

Gmax

(0 + -) cos 0 D i
G t - for 00 WCO

(4. 5-52)

'A

Therefore, at X f+ I
k~m i

max

G max p G

D for = 011 Lfor0

XA+ 1

x-Lfor max D

0o

180o

XPd-l1

1 f ~1,B0"10

For Oo = 900 or 2700, Eq. (4, 5-50) may be solved for X, giving the following equation: V1 From Eq. (4.5-40): +V(4.5-53)

max

sin D(2 D)

(45-54)

The value of G is the same for k. The sign in Eq. (4. 5-54) corresponds to the D term In Eq. (4. 5-53), Therefore for 0 = 900, Gm wx 1/2D at X u Y1=, and Gmax - IY2D at X = -. For 00 270*, Gmax el1/2D at X= V= and Umax l/2D at X V =T'-. Note that, for a given Fo and r (i, e., Omax), the potential . '-11 torque for an in plane asymmetry Is approximately twice that of the out of plane asymmetry.

-184 -

14 1

SI.',

Sll

'I
?"t JOHNI
MOPKINI

UNIVIWtIITV

APPLIED PHYSICS LABORATORY


N UILAVINSWll k,N MA*VI.AwDJl

7 ..

For the general value of 00, the maximum valuns of G may be obtained from a plot of F(X, D), Since ~F(-X,, D) - -F(X, D), It is necessary to plot only the +X. values of the function, The maximum values of G for +X are obtained at MF(, D) a tan 0o; the values of -X are obtained at F(X, I)) - tan Oo. A sumnmary of the characteristics o(f G is given in the table for in plane and out of plane asymnmetries. 01
QM
Poi/v

01 .

.
Umak
2 D i

, MAX
11 4.5

270. x 0 MA

maxx

"rn

Negativye

-I/il

I -*1+-.-

-1/2D]

t4TW"

.lID

ID t./I2D1

I,

0
Unatabil'

*t
Unatable at +1 nStable at -1

(3

tabllt l S

St ble

tAbl

It 4+1

saltable

at "I

Typical plots of G versus X for D n 0. 2 are shown for 00 - 0*, 450, and 900 in Fig, 4-15, The values of G at a given X for 00 = 1 800 and 2700 are the negatives 00 O0 and 90% respectively. Note that of the values for for in plane asymmetries the roll torque from combined asymmetries is small except for a narrow band of X in the vicinity of 1 (i. e., near resonance conditions). However, for out of plane asymmetries the roll torque may be appreciable for all values of X between about -1. 5 to 1. 5 except at X 1. The function F(X, D) is shown for D = 0. 2 in Fig. 4-16. Using Eq. (4. 5-50) as the condition for Gmax, note that, except for small values of 0o (less than 00 - 45), X- 1. Furall values of Gmax tend to be nearX = 0orX thermore the magnitude of Gm x tends to be small when it occurs at values of X other Ean 1, For example, the maxima for 00 450 or F(X, D) = I are shown in Fig,

-185-

1
'I 1"..,-.,,.,,'.. .

U4 JOHN$M H@Oo4INS UNIVIRNIS APr,.IK6 PHYSICS LABORATORY


mli.1* 11ING MARV6AND

4-16 to occur at X-" 0,12, 0.'74, and 1,06. From Fig. 4-15 the corresponding values of Gmax are -0. 7, -0, 9, and 4.1, respectively, 4. 5. 5 Roll Resonance One of the potential problems resulting from asymnietries is prolonged periods of roll resonance. During this period the angle of attack increases, resulting in large lateral acceleration and high drag, If the resonance occurs during the heating portion of the trajectory, the problem may be amplified byjdistortion resulting from asymmetric heating since at a one ray along the body remains on the windward side of the body and is heated more than a corresponding ray on the leeward side, In this section we will consider conditions under which prolonged periods of resonance may occur. The term 'lock-in" will be used to describe a roll history in which Pr tends to follow WA so that X f constant. That is, 1 -. It should be noted that this does not. necesS/1/A sarily imply that resonance (X - :hl) exists. However, asymmetries are usually small enough that roll amplificration is required to cause significant changes in roll rate and the necessary amplification can occur only for conditions at or near resonance. Typical values of roll rate at reentry are I to 2 Hz, and the static margin typically Le 3% to 476 of body length. For these conditions, the peak value of WA is much greater than PE so that X = I will occur at least once during reentry. If the range is short enough or the reentry angle shallow enough, the resonance condition may occur a second time, as shown in the sketch on page 187.

4,

iI
. . .

I
ThE J@MNUlM@PWINUluNiygRU'ITr

APPLIRD PHYSICS LABORATORY

FIRSTN
RESONANCE

SECOND
RESONANCE

W-A

ii
"1

A vehicle having no roll torque would exhibit a reduced roll rate (Pr) trace shown by 1.. The roll rate would

pass through "first' and "second' resonance without incident. However, if asymmetrie~s exist, the roll rate may
Curve 2 is typical lock-in on WA a shown by 2 through 5.

~of the case of large asymmetries that exist early in the reentry phase. in this case lock-in may occur at "'first" 1n the more usual case, asymmetries develop resonance. I'' as a result of the heating environment so that "spin-up" "first" (cases 3 and 4) or "spin-down" (case 5) occurs after S. resonance. For cases 3 and 4 the angle 00 is between 1800 and 3600 so that "spin up" occurs; for case 5, 0 is - between - . 00 and 180* so that "spin down" occurs. Cases ?and 4 indiCate that p may lock-in to WA (case 4) or may cut through the WA curve (case 3),
..

~.
...-

I..

The roll lock-in conditions are the same whether the vehicle has positive or negative roll rate at reentry, -187-K

THE

JOH146 MOPKINU UNIVERSIY


OILVIR Op"ie MNAgRVI.ANU

APPLIED PHYSICS LA9O01ATOCY

To simplify the presentaiion, only positive roll rate will be considered In the remainder of this discussion. If the reduced roll acceleration (Or), produced by asymmetries, is equal to or greater than WA, lock-in may occur. In equation form, lock-in may occur when:
._

(4.5-55)

ly
Using Eqs. (4.4-37) and (4. 3-18), it may be shown that 4A is given by the equation (neglecting variations in temperature with altitude):

FC
WAVE2

&Xm sin 3 y12 N8 CAY l

,K

Using Eqs. (4. 3-8), (4. 5-36), and (4. 5-38), f~Is given by
the equation: G VE M
2

m sinCN " r

K '

2 IxCA

(4.5-57)

Therefore, lock-in may occur if the following inequality is satisfied: L- G ax 2 CN H m r2


___

od o-

I
yd l

siny
Iytr

121

/ ,,

(4.5-58)

"IN

Equation (4. 5-58) may be simplified further by noting that, for a given altitude (or given value of K), the minimum

L,

-188-

-. -. ---------.-

rum ~~ ~

.-

~.,.nI~lIWflln ~~~~~.

...........

APPLIED PHYSICS LABUOATOlRY


vE iPL A 6091pi. MAMYLAND

HRSJOHNS HOPKINS uNIVRMSI V

asymmetry that may result in lock-in is given by Eq. ax. If trim conditions are realized (4,5-58) for G " is given approximately by G 10)5then 4.;, Section (see (see Eqs. (4. 5-52) and (4.5- 4j): Gmax =k/D where k has the following values: (4.5-59)

0
900
900
90

I
12

1 +2

-0. 5

1 2

-l

1920 80

"1 1+1 -- "0,5

2700

1 - -2
2

0.5

UsingEq. (4. 5-20) and the definitions of F. and Mq, an expression for D Is:

D
where

k1

Cmq]

(4. 5-60)

k, =

0 (

'y
I

ymd Y
Substituting Eqs. (4. 5-59) and (4. 5-60) into (4. 5-58), and noting from previous discussions that Gmax occurs near the resonance condition for in plane or out of plane asymmetries: r R (4.5,-61)

-18 9-

I
LL

H
7',

FT7
?NU JW4,,NI M,OWKIF4 uNIV II|tNBiI

APPLIED PHYSICS LABORATORY


6I1.VIR OPRING. MA1ILAND

where

F1 =

/I-

- c

(4.5-62)

and KR is the value of K at the re,,onance condition, The general behavior of Eq. (4.5-61) is shown in the sketch,

4MAXIMUM

ad 2

ItMUST ItS

*)

"-W

k M UL ST 6 E

-';

'

LOCK-IN POSSIBLE

NOT POBSLK

NO

POSIL

NO

KOS OLE

REENTRY

d-P
- 19 0 -

APPLIEO PHYGICS LABORATORY

TIM9

HOPllUMS JOW4NS

UNIV&PIT'Y

Following reentry (K

0), aSSUmne that resonance (G If o r/d

Gmax) occurs at an altitude where KH = K 1 ,

(0

r/d)1 , lock-in is likely to occur at first resonance pro-

vided k> 0 and lock-in will continue at least to K = I unless dynamic effects (roll raWte nd n oscillations) or roll damp3reak ing effects (sec Section 4. 5. 6) cause "break out."

out will occur at Klj = 1 unless 00 is such that and therefore G change sign near X 1. Therefore, from
Fig. 4-15, break out will occur at KR = I for in plane asymmetry (00 = 00 or 1800) but may not occur for out of Break out would plane asymmetries (00 = 0or 2700). occur, however, even for the out of plane asymmetry at K 3 . To be assured of freedom from lock in at all

conditic'ns: or (ry )

whichever is the more restrictive. The subscript 2 refers to the second resonance condition. Neglecting a change in roll rate during reentry, it Since the slope may be shown that 11 - K 21I I - K I. of the negative the of the ( r/d curve for K = 0 to 1 is more the slope for K > l, the (O r/d)1 limit is always resonance the fI'om Therefore, restrictive condition. standpoint, it is desirable to select a low value of PE so that first resonance occurs at as high an altitude as possible. The maximum asymmetry allowable is obtained by evaluating Eq. (4. 5-61) at K = 0 so that to avoid lock in:
-r
<--

(4.5-63)

In Eq. (4. 5-61), the asymmetries are given in The ar':ymmetric moment coefficient terms of e and r. is a better parameter for some purposes since the moment coefficient is a direct function of the asymmetry whereas Co depends also upon CN, and vehicle stability (AX). The components of a are shown in the sket'ch on page 192.

.4.

"-.

,Ik

"

"'

'

"

THR JOHNI

HOPKINS UNIVRRSITV

APPLIED PHYSICS LABORATORY

IlL.VIIII~

~~

0PhaMI.N

SIN
CASE FOR 00 900IS SHOWN

01
zI
The component of a resulting from the center of gravity offset is always directed away from the center of gravity, whereas the component resulting from asymmetric moments may be in any plane. The components of a are (see Eqs. (4. 5-27), (4. 5-28), and (4. 5-29)): Cr CA dx
CN

-o

~c.g.

(4, 5-64)

Nd

ax
=o

Cm
N d

(4. 5-65)

where
=

--

Cm

n2

(4.5-66)

-192-

, . ,.

., .

I
?149 JOMN% HOPKING UNIVIRIYTV

APPLIED PHYSICS LAB3RATORY


* p

MAELbA4o fllWkRA opm

For

toward the center oi OM must be directed Then>&T. gravity and nOM >Ca . o00o
-r

SA(4.5-67)

Cd
For 0o =180: -+ C.A.

m
a . AX

r Ad

For o

=900 or 270*:

"21

ag <<y OM

so that Eq. (4.5-69) is given ap-69)

proximately by the ,xpressiton:

[c

Usua ll

C CN mo

9.

(4.s -60) v

"

Substituting for;o, Eq. (4. 5-61) becomes:

(i-K
F A< F
2

(4. 5-71)

"-193

+i)
.-

TH'K JONNH HOPKINS UNIVKMMITV

APPLIEO PHYSICS LABORATORY


5lLa1 SPUI~O
MANYLAANO

where F
r, A

CC

r m~ 0 d

1C

CA
m
0

r-

fcr
'0 0

0
0

dor
,
00

A
Cm o

(4.5-72)

C 2
-1
0

for

900 or 2700.

F 2 =C N F I*

(4. 5-73)

The Iimitirng asymmetry values may be obtained for any in plane or out of plane asymmetry using Eq. (4, 5-71) with the appropriate value of k given in the

table in Section 4. 5. 5. Note however that (I - KR)/k must be positive.


The possibilities for lock-in rre shown using the roll amplification diagrams shown ci page 195. At rcentry, for a positive roll, ). + so that G - 0 and the roll rate remains constant. As the altitude decreases, WA increases and therefore X decreases. As

X -. 1, G is no longer negligible and, if the f'r produced at pointo 6 (0 = 00, 90% or 2700) is equal to or greater than (;A, lock-in may oc.ur at some value of X between
a and b where G is just large enough to satisfy Eq.

(4. 5-55).

As the maxir urn dynamic pressure (K - .) is

approached, (A decreases, becoming zero at K = 1, and then it becomes negative. As a result p must reverse 00 nor 00 900 sign if lock-in continues. Neither the 0 00 asymmetry case can produce the characteristics required

-194-

-- . -+-.. . +-, -

. u q nrol

impM

THEl~JOHN$l HOPKINSl IJNIV9MIIII

APPLIED PHYSICS LABORATORY


SILVIA 9PRIM43 MARYLAND

00.90

0 / , .L

RATIO OF IMIDUClID 14OLL FIATC

tt,

TO PITCH FMIOUlINCYv'."

IsD
I
-i

"*for
1195

00 = 2700 until the negative required for lock-in is

greater than that given by point c. For the 00 =180" case, lock-in cannot occur since k (or Gmax) is < 0, The vehicle would pass through first resonance with a slight reduction in roll rate, Both and eotend to cause a rapid transit (both effects tend to reduceAX) through the reso$I nance condition. As WA increases, X decreases to some low value and p remains nearly constant. However as Wdecreases at K > 1, A. increases. Lock-in at this time (second resonance) requires a negative and may occur between points b and c if at c is large enough. For Oo 900, the shape of the G curve is such that lock-in

YT1i JOWINB WOPKING UNIVKRSIYl

APPLIED PHYSICS LABORATORY


916V#1 OPRINS MARYLAND

could occur at second resonance between b and c. However, if the torque at c is large enough to cause lock-in

and if the asymmetry existed at reentry, then lock-in


would have occurred between a and b at first resonance, and X would never be less than 1 + D. Therefore lockin at second resonance would not occur. If, on the other hand, the asymmetry developed during reentry and after first resonance, then second resonance lock-in for 00 900 is possible, Note that lock-in may occur only when dG/dk. is negative since, for positive slopes, X is unstable for reasons discussed for Xe stability in Section 4. 5. 4. The most restrictive limitation occurs for first resonance lock-in with the in plane asymmetry with Oo = T. For this case, with first resonance near K = 0, if lock-in is to be avoided:

.I

CA
0

r < F2,KX Ud'

(4.5-74) -

For _ A-- E <<1:

"<

r < dF .2 T

t'.5-75)

From Eqs. (4. 5-62), (4. 5-73), and (4. 5-75), note that the allowable asymmetry is governed by the following parameters:

I'

II
_A.I:

-196-

twa JOHNNHOPKINS UNIVeSPITY

APPLIED PHYSICS LABORATORY


MILVIINOPOING MIAMYANO

Cm r to avoid lock-in

Parameter Vehicle stability margin Resonance altitude Planet of asylmmetriesa Aerodynamic coefficients Type of trajectory A tmosphere Vehicle slenderness Mass distribution factor Vehicie size

Related Factor A&X/d 1 - KR k

tends to be low for: Small ax/d 1 01 or 1M06 (compared to 906 or 2700)

CN and C N m q
sin y
H-

Low CN and Cm
Low y (shallow reentry angles)
High H

IX Iy n C m I,, /md d

Low Y,'

a"~r Cn

Low 1y/md2 Low d

Since there are so many variables, it is not possible to present generalized curves of allowable asymmetry. However several observations are noteworthy. Consider a family of bodies differing only by a scale factor. Values Of CN, ~Cmq. IX /Iy, and Iy/md 2 will be approximately the7 same for all bodies, If it is further specified that all bodies have the same stability margin (same AX/d), then for a given atmosphere and trajectory, Eqs. (4. 5-62), (4. 5-73), and (4, 5-75) show that the allowable CM r is proportional to d , Therefore, small bodies may be particularly sensitive to mass and aerodynamic asymmetries. Note also that for a given allowable F 2700. For 00 .00, the minimum occurs at:

"Em 0 (Eq. (4, 5-72V has a minimum value for 4.0.

the allowable 900, or

d2 CA
AA

APPLIED PHYSICS LABORATORY


6'I.kIL SPIVlNG MAP-LANO

TMI9 JOHNS

HOPIINS UNIVZISITV

and the minimum Cm Is given by:


0

Co<2
For Oo

Ad 0

GF 1 - K,

(4.5-76)

90* or 2700, the minimum occurs at:


0C

<m

rm /

(4.5-77)

and the minimum value is given by:

c CE

lr0

< 2 29 CA

2 (1

-KR)].

(4,5-78)

Therefore for a fixed value of FA the minimum value for "mCo for an out of plane asymmetry is about 15% greater than that for an in plane asymmetry.

4, 5. 6 Effect on Lock-in of Roll Damping and Pure Roll Torque


The roll damping, M., and pure asymmetric roll

torque, lob terms in Eq, (4. 5-33) are usually negligible.


However, under some conditions these terms could be significant. For example, at usual roll rates M is small, p but if lock-in occurs and drives the roll rate to high values, M may become a significant torque and may, in fact,

break the lock-in condition. terms:


GI

Rewriting Eq. (4. 5-36) to include the M

and I

p
(4.5-79)

max

-198-

THl JOHNS HOlPNIN$ UNIVERtSITY

~Ii

APPLIED PHYSICS LABORATORY


lt* *ORIN* MAIRVL&NO

"where
G'
G +G 1 +G G2 (4. 5-80)

The amplification factor G i defined by Eq. (4. 5-37). factor G, is defined by the equation: Mpp max
V

The

'd

CN

Vd

Using Eq. (4. 5-21) to eliminate p: cII

c AdX

1cN C01

0 "I--y

and using Eq. (4. 3-23) for w0:i

[-AXK sinT]/
GI 1 2HO A k I-'r

CY
)'=a).

12'1
, (4.5-82)

The value of a is always negative. Therefore if the G curve is drawn, the GI = G + G, (assuming G2 : 0) values are the ordinates measured from a X1 axis which is rotated counterclockwise through an angle, c, where: tan( =a -199r,, I

(4.5-83)

I,

?Nk JONNS UOPKINS UNIVkelyV

APPLIKD PHYSICA LABORATORY

0
I

Similarly: C G2
0o 0

0oa

C-

(4. 5-84)

Xmax

od

If G 0, thL values of G' r `, + 2 are the ordinates of the G curve measured from a X axis that is shifted by
C

an amount C.

as sketched (for a negative vralue of

-I'
-200-

..

-..

'-

* ....*?.,

-..

*." "'*~'

" :"

d'

:'"

"

'"

"

jIAPPLIED

ypqg PHYSICS LABORATORY


J014PIN MOP0INU

tWNIVRRSITY

IILVIII SPMING. MARYL.AND

Suppose an in plane asymmetry exists and ao and r/d are such that an ampli.fication factor, G a GR (see Eq. (4. 5-58A, could result in a lock-in condition at first resonance if CI 2 0, Lock-in may then occur at point A in the sketch.p

G~ G

C I

e2

-201

I
I

"TP41
JOHN*

MOPKINB UNIVIRpITY

APPLIED PHYSICS LABORATORY


91.VIn l SPRiNG MARVLAND

However, suppose CIp

0 so that e for resonance at the

initial roll rate is given by (.

The resonance condition

then is obtained by drawing the line BC as shown, giving a lock-in condition at B. As resonance continues, K increases and therefore C increases, If ( exceeds the value shown by (2, the torque required to sustain resonance is unavailable and break out occurs.

tan

C *

Using Eqs. (4, 5-58) and (4. 5-60), and noting that R -max -G G "aG R Gmpx - GR' it may be shown that roll

max damping will cause break out when the following inequality is satisfied:
2 (H/d)CAklv
-r

k
KR

CA
md2

((.5-85)
J

" Cmq

sin (-,y)

Roll damping tends to decrease the possibility of lock-in at first resonance and, if lock-in occurs, tends to increase the possibility of break out. The opposite effects occur at second resonance. When CIO exists (but not C) lock-in cannot exist

unless

UiG Eq. max

GR

> >0 -ro CN _an

1
T

Using Eqs. (4. 5-58) and (4. 5-60), lock-in cannot occur at
first resonance if Cp has a sign opposite to that of the initial roll rate and a magntude given by the following inequality:

I, ,.

-202-

APPLIED PHYSICS LABORATORY


li*'Vii gPftIN.i MAamN O
-'N

1141 JOHNS HOPKINS UNIV9""Iry

|
~

0.

t -r

1 od

R R
a----1 K

,K

(4,.5-8(3)

where

CR

> 0

. N k = 1 Cnnq

CAX k

kC

sin (-)

(4.5-87)

a2

1
yI~

(4 5-88),

If the sign of CIO is the same as that of the initial, roll rate, the effect of CIO is to increase the probability of lock-in; if CI and PE have opposite signs, the effect is to decrease the probability of lock-in. 4. 5. 7 Spin Through Zero Roll Rate

In the previous sections, the problem of roll resonance was considered. A second problem resulting from variations of roll rate during reentry occurs when the roll rate becomes equal to or passes through zero. When p = 0 and an asymmetry, go, exists, the lateral acceleration resulting from is fixed in space and the missile moves i0 off course in the direction of Foe The impact dispersion produced may be much larger than that required for satipfactory performance. In this section we will consider the conditions for which spin through zero may occur. Referring to curve 5 in the first sketch in Section spin down through zero roll rate may occur when

4. 5. 5,

-203-

APPLIED PHYSICS,, /ABOOATO3R .


SiILVIEI
*IPltW3

TNC ,JOHNSI H'OPIKINSI UINftIV|!IY M.fAI'LkD

I!
is in a ThereThe enough

an out of plane component of asymmetry eAlpts and direction to reduce the magnitude of the roll rate. fore, one condition for p = 0 is that 180 " 00 > 00. second condition is that the magnitude of is la: ge to result in p = 0 prior to imprict.

For the asymrnmetries of interesi in this problem, the vehicle usually passes through first resonance with little change In roll 'ate so that mrnt of the change in roll rate occurs at ), noar zero so that roll amplification is negligible, The roll acceleration, then, is given by Eq.
(4. 5w-35):

-CN "qs tor sin 0


N 0
-

.(4.

5-89)

Ix For very sreail angles of attack (eo):


q C C

.x

1 dV .g dt

CA S W

Therefore:
" -d" qS A

"-

(4. 5-90)

Substituting Eq. (4. 5-90) into (4. 5-B9):


dt

CN

d0

r sino

0n m

dV dV

(4.5-91)

A .
i

If p and V ace assumed to be the only dependent varla.. blE.s, Eq, (4. 5-91) may be integrated to give: CN P P
A p

rr sina 0

roVE( '

).

(4. 5-92)

-204 -

Ii

APL ED PHYS:C.S

TImS Jflp4Ns moofos

LABORATJI',

up4..vfRr,

Substituting for V/VE using Eq. (4. 3-5): ']

~
=

~CNo
-

i
d V0E -K/2
-

4p 1 C

(4. 5-f)3)

A X

(IN/rod() at I<
1(

If

LAP

> P

<SL'

spin through zero will occur prior

to impact. %rhealtitude (or K) at which zero roll rate occurs may be obtained from Eq. (4. 5-93):

PE A m2

Pc

IX

e =1 /2 -

N(4.5-94)

(Y

sin~ ,
1 (4.5-95)

or

K =-2 in

I CN

PEA md 2 ,
Ci
-- r VTE

sin

L N

ad

The minimuin asymmetries which may cause zero roll rate occur for 0) = 900. For this case, Eq. (4. 5-95) may be written in terms of the moment coefficient using Ea. (4. 5-70):

In

I -

PECA C~ m2 P m-2

(4. 5-96)
d-

o -d Mn

L2

m
0

If spin through zero is to be avoided for the entire trajectory, Eqs. (4. 5-95) and (4. 5-96) indicate that the following inequalities must be satisfied:
-

205

tME .I,)MNS JPIRINI

LJNIVE.SIIr'

APPI-ED PHYSICS LABORATORY


*,LV* v41plllPU. MAAIU1huI3

E Vrd 0o 1 p

A
2

md

CN

( PS.L.)

(4.5-9 7)

s in_ ,

or In terms of C

FA

oA *

1 (CA2 C (C m 0

r2 d)2

lVE X(:.PsL.) An-d(.-8


x (Ci2, sin

(4.5a-98)

Using the assumptions listed in Section 4. 5. 5, pale 197,


note that the allowable m, r is proportional to d . Thev'efore, as ir, the case of the roll resonance problem, small vehicles are particularly susceptible to the spin through zero problem. The probability of encountering the spin through zero problem on a given vehicle is highly dependent upon the missile-to-missile variation in 0.. If the asymmetries are such that 00 is always 900, and if Ko r/d always exceeds the value given by Eq. (4. 5-97), then all the missiles will spin through zero. On the other hand, if 00 always lies between 180* and 3600, the probability of encountering spin through zero is zero regardless of the value ofo r/d. If the missile-to-missile variation in 00 is random, the probability of encountering spin through is given by:

S~zero

ff

-206

Tht

JCH4NS HOPKIP4 NIVU4diVK5Y

APPLIED PHYSICS LABORATORY


3ILVWI IIPlNB
MA&ILANO

II
A=cos "
A[
0o

where A is an angle between 0 and f /2 radians defined by: "


n , r1i

r/d

and (go r/d)rnin is the value of &0 r/d given by Eq. (4. 5-97). For this assumption, the probability of encountering spin through zero varies from 0 at io r/d = (o r/d)min to 50% at ;'o r/d :i (0. r/d)min' 4. 5. B Effect on Products of Inertia The products of inertia are defined by the following equations: IXY I I mXY MMm X Z (4.5-99) (4.5-100) (4.5-101)

i ii Yz =EmYZ

The IyZ term is negligible for most reentry bodies and is omitted from this analysis. If Ixy and IXZ are not zero, the principal axes pass through the center of gravity but are rotated through an angle 8 measured clockwise from the body centerline. For small asymmetries, the rotation angles are (for Iy = IXy
Iy

(4.5-102) x

SI

"" y

I(4.5-103) X

*1

.
-207-

APPLIED PHYSICS LABORATORY


S-lS SPRINO MARYLAND

t;W

JOHN% HOPKNIN UNIVV5IISt

(4.5-104)
where 8 about the is a rotation about the Z Y axis, 0 but axis and 6 is a rotation

IXy.

Consider a mass asymmetry in which IXz 0, as illustrated by the sketch.

/1

Assume the vehicle is symmetrical except for two small masses, m 1 , located at (X 1 , Y 1 ) and (-X 1 , -Yi) so that S=2mlX 1 Y 1 , If the missile is now rotated about the X axis, the centrifugal force acting on each m 1 produces a yawing moment about the Z axis. Returning to the case where Ixy 0, I # 0, and the mass asymmetry forces are much greaterthan the aerodynamic forces (X -. cc), and retaining only the larger asymmetry terms (p is moderately large but 1Xy, 1Xz, q, r, , t, and tarc assumed to be sufficiently small that the square or product of two small terms may be neglected) of Eqs. (4. 5-1), (4. 5-2), and (4. 5-3), the equations of motion become: -208-

S.

... , ..;,-!..,1 'l,,',i.1 ':

1
TI JOiNG HOPKINS UNIVIENIIITY

APPLIED PHYSICS LABORATORY


SPRtING MANUANO

SIl.VUR

-0 2 IXZ =rwg - p

(4. 5-105)
(4.5-106)

2 IXy y
g P (4.5-107)

where w

= P11

--

j.

Differentiating Eq. (4. 5-106):

S=

tf.

(4.5-108)

Using Eqs. (4. 5-107) and (4. 5-108): +Wg 2 q =gP 2 Xy IY (4.5-109)

The solution to this differential equation is: Iy


q=C

sinw t + C2 cos W 9t+ p

j
(4.5-111) 4.5-112)
(4. 5112)[

S Wg cos Wgt - C W sin w

,t

From Eqs. (4. 5-106) and (4, 5-ill):


r

C 1 cos Wgt - C2 sin wgt + p y


9 2 09 1

XZ

-209'r )1

THE JO00416 iPoPKINm UNIVINgITY

APPLIED PHYSICS LA1ORATORY


*ILvI[g mpal... MARYL.ANO

From Eqs. (4. 5-110) and (4.5-112):

q p

+ r( p

-'x

Therefore, the q - r plot is a circle of radius with center at q


=

+C

p I IXY

In other words,

the telemetry traces for a body having nonzero IXy and IXZ show a q which oscillates about the value q

the~ teeer

rpY

I
y

and an r

which oscillates about r

p I

The con-

stants C 1 and C 2 may be evaluated from given conditions a rt t 0 ... IXZ :

CI

tP PyI

-X

(4.5-113)

I C2 :qi P.Iy-

(4.5-114)

Equations (4. 5-113) and (4. 5-114) may be used to


evaluate Ix -and IXZ from flight test data if p, q, and r

are known accurately.


In space, where the aerodynamic forces are zero, the principal X axis moves along a cone. The cone half-' angle is given b-/ the equation (see Eq. 4.4-23):

tan 0

(4. 5-115)

I
-210-

.. 1

I
T""r JOMNS HoPSINi UNIVgRh1?V APPLJED PHYSICS LABORATORY 0P0.NG MARN6AND *1Lw.V

The period of the rotation about the space cone is given by Eq. (4. 4-25). Since the body centerline rolls around the princLpal axis, the angle of attack oacillat('.; -ith a relatively long period (%-r) and a peak-to-peak amrlcude of 20 with a superimposed short period (l/p) oscillation with peak-to-peak amplitude of 26. The constant shift in q and/or r resulting from nonzero products of inertia may be interpreted in terms of angle of attack using the following equations (see Ref. 4-9),

"1

*q

)I
*

'-p

1y

XY i Xz

X-

(4,5-116)

(4.5-117)

When the missile descends to altitudes where the

aerodynarnic forces are no longer negligible, assuming that tyo and A. may exist as well as ab and b the total
angle of attack is given by the equation (see Ref. 4-9):

A (

x2

2 + (e5 + +2)2]

(4.5-11 )

where A is defined by Eq. (4. 5-16). trim force is defined by the equation:

The direction of the

0
where

00+ A

(4.5-119)

"tan
0

(4.5-120)
o0.

ii

if

and

tan -211-

(4. 5-121)

?TH JOHN% HOiNtING UNIVeIRIY

APPLIED PHYSICS LABORATORY

~Ii

S.Vili SPRgI. MASYI.AND

ifa CO

0, and

-N

Ktani(.
or

2~+j,

tan'

a
(4.5-122)

Ai
if either

=tan'
and 13 or ei

D 2
and 0.

The proper quadrant for 00 may be obtained from the signs of the numerator and denominator in Eq. (4. 5-120). For 0o - go = 0, the trim angle of attack is just that resulting from products of inertia. Let

%=

2(4.5-123)
is given by: (4. 5-1 24)

Then, the amplification factor in terms of Amplification factor =

* Iil
-212 -

APPL.lED PlHYSICS LABORATORY

9II JON1491MlOPKINC LJNIV91401II"

*1
The amplification factor ts zero nt ), 0, I-as the same amplification as that for aerodynamic trim at X = I (namely, A), and approaches unity as X-. -. Therefore the products of inertin may be important at X near or grenter than unity and, Ln combination with the center of giavity offset, could result in resonance lock-in. The effect may either augment or diminish the effect of other asyrninetrius, depending tpon the signs and magnitudes (see discussion below) of Ixy and IX . However the product of inertia asymmetry in combination with the centcr of gravity offset cannot result In spin through zero since the roll torque rt p = 0 is zero, Consider a vehicle having a positive roll rate at reentry and a combined in-plane asymmetry consisting of a center of gravity offset along the +Y axis and a positive product of inertia IXy. At reentry (X- w) a positive 1Xy results in a positive g (Eq. (4. 5-116))or a force F,, along the -Y axis. Since F rotates clockwise as X decreases, F is always negative and the roll acceleration resulting from F is always positive, and therefore lock-in at first resonance is a possibility.

FtF

Xf ,

/:
!!/
-

/!
Z

23- -

THEr J014NS )'OOKlL41 UNIVIrI'IIIITV

APPLIED PHYSICS LABORATORY


OILV1RSPlINO SA.fl, r.ao

Solving Eqs, (4. 5-1) and (4. 5-2) simultaneously and retaining just the b contributions resulting ft-om aerodynamic forces:

I
x

+ XY m
I I and y m using Eqs, (4, 5-4) and (4. 5-5):

Substituting for

Fy Z c.g.

F AX I Z XY

Ix

y-

Note that fbresulting from the second term on the right side of the equation is negative since Fz is always negative. had been selected Fz would Similarly, if a negative Iy always be positive. Ther efore this term results in a negaand, for positive roll tive P regardless of the sign of Iy rates at reentry, tends to prohibit first resonance lock-in but tends to cause second resonance lock-in, In a manner similar to that used in deriving the criterion for lock-in for an aerodynamic asymmetry and a center of gravity offs'ct (Eq. (4, 5-61)), the criterion for lock-in with a product of inertia and a center of grRvity offset is: r
-2

IX

'

This inequality is identical to the inequality for an aerodynamic awymnietry and a center of gravity offset (Eq. (4. 5-61)) only for the particular case of a neutrally stable vehicle (AX/d - 0), which is not of practical interest. For nny staticall stabie vehicle (positive AX/d), j r/d is :tJwaj;i greater than eo r/d for a given F 1 and KR. Solving for rY 8 at the boundary (the above equation set to zero) using the binomial theorem:

-214-

1
___

__

APPLIED PIYSICS

ASAB0RATnRPe

(~":~) ~

'411(F

)2j

(4.J5-125)

where

11

F4=2F IF3 Note that lock-in cannot occur unless:

For the mninimum value of F 3 r/d, 6F


33

F4 1=1

KRi

Another lock-in requirement is that:

8 a min
where an and

66 max
8

are given by Eq. (4. 5-125) using

the negative and positive signs of the radical term, respectively. The general requirements for lock-in for this type of" combined asymmetry are as shown in the sketch for a particular value of K

:*

215

"I!!

T71l JON1eI HOPRIND UNwV'RlTYT

APPLIED PHYSICS I.ABORATqRY


SIL-10 SPNINU MAOYLANfl

LA6

lASYMIPTOTE

LOCK-IN NOT POSSIBLE

LOCK-INLOCUS OF

r ,., (Fo
a6

AT 7i4

ASYMPTOTE

0
3

( m

,n

KR I

F3

"'

Note that reet r id)t decre y sses from t7 -at reentry (KR. = 0) to 0 at maximutii dynni-ic pressure (KR and then increases to/"' I1-K . i-. <t impact.

.', ;'. [

Current

reentry vehicles are designed sulch thant first resonance occurs at Klit 0, so that I Klt 1. Assurniing p does. not change during reentry, it may be shown that II - K~t >1 at second resonance. Therefore lock-in is not possible at first or second resonance for any value of '6 if the vehicle can be constructed so that:
-216-

1
"1t

______

APPLIED PHYbVC S LABO)RATORY

V-

JOH-6NS it,"*SU O

____7_P_

24T

SL~~i

LUtL~

fo (

VI' a Id F3thi's

inequal~ity b~ecomes

genstcI/& tin~~
X Y
q

si

1/2

*'

tedt

be inldependent of size.

Then, for a given shape and AX/cl,

the allowable center of gravity offset varies as d32and [sin (--/)1 1/2 Therefore, small bodies flying shallowangle trajectories are more susceptible to lock-in than are large bodies flying steep-angle trajectories. 4. 5. 9 Unequal Pitch and Yaw Moment of Inertia For a body whose only asymmetry is unequal pitch and yaw moment of inertia, the equations of motion (neglecting sn-iall terms) become: QY -II Z)qr (4, ~-12 6)

= mn Ni

-pr(IO

(4,5-127)I I )
-

=n -pq(I Y

4 518

1)qr term in Eq. The significance of the (l sketch. The body in the Iz > IY) shown (for 5-1 26) is (4 snmall eq~ual masses for two except symmetry has mass from the center (AY) distant equally axiE Y on the located of gravity. At some instant, the bod4y has t. transverse rate of rotation shown by the vector T At this time the

THE
JONNIN

-OPKINS UNIV9I.mI'Tv

APPLIED PHYSICS MA*,LSNO LABIORATORY ISLvt. Set.

rn

II

r
t

CF" CENTRIFUGAL FORCE

body rotates about an axis coincident with T and the centrifu al force acting on mI produces a couple -2mlAY qr = (ty - IZ)qr about the X axis. The roll torque is negative when IT lies in quadrants I and III and positive whenT lies in quLadrants II and IV. At very high altitudes (X- m), the terms m and n in Eqs. (4. 5-127) and (4. 5-128) are negligible and the equations for q and r are those given by Eqs. (4.4-18) and (4.4-19): LT Cos , t (4, 5-129) (4. 5-130)

WT sin w t

At t; 0, C T lies along the positive Y axis and, for positive p, rotates counterclockwise with constant amplitude the asymand frequency as t increases. Therefore, occur in roll rate metric mass and the rotation of WT result in a roll rate oscillation of cycles Two oscillation. for each revolution of WT about the Z axis. The equation for the roll rate history is obtained by substituting Eqs. (4. 5-129) and (4. 5-130) into (4. 5-126) and integrating to give:

- -.

, ....

2..

,'1

I
t~ 01M HOPRIP46 UNIVEM6ITV

APPLIED PHYSICS LABORATORY

"_"_

- sin wait

(4,5-131)

where

gl Pt1

('

Therefore, the roll rate trace in the exoatmosphere oscillates as a squared sine function with a peak-to-peak amplitude of:

2p P\Iy

2 2 Iy - I(Z

Ix/lx

The frequency of the oscillation is

wgi.

When the vehi.cle descends to altitudes where aerodynamic forces are no longer negligible, the ZT develops a pattern of motion that depends upon the initial conditions and the vehicle's aerodynamic characteristics, Unless the rotation of ZT about the Z axis becomes very slow, a net change in the magnitude of p is unlikely and the effect on the reentry trajectory is rnegligible. Under some conditions it has been found (Ref. 4-10) that the direction of rotation T about the Z axis actually changed directions during reentry. During the period when the rotation changed sign, a significant reduction in roll rate occurred. The possibility of encountering difficulty as a resuit of unequal Iy and IZ appears to be remote, and this type of asymmetry is not considered further in this report.

-219-

THE JOHNS HORSINI UNIVSRSITY APPLIED PHYSICS LABORATORY


SEILVERSI NSlll MAILAND

4.5.10 Applicability of Trim Characteristics at Resonance In preceding subsections, it was assumed that trim conditions are nttained at first resonance, However the resonance condition is approached very rapidly, and large ciLnnges in trim occur. Unless the missile can respond very quickly to the changing trim connuitions, the missile
will pass through resonance without attaining large ampli.-

SWA.

fications in the angle of attack. The weathercock frequency, is an indication of the ability of the missile to respond to a change in trim. IfWA is high the missile response time is short; if WA is low the missile response time is long, Since

*1

P
WAy

AX4S
x1

WA is proportional to 't Therefore, no matter how much static stability (AX) the missile has, the response time is always long at very high altitudes. If resonance occurs at a sufficiently high altitude, trim conditions are not attained and the loads and lock-in criteria based on trim equations are not valid, The conditions for which trim theory is valid have not been examined in the literature. In order to provide some insight into this problem, a brief study was made of the lock-in criteria using a six-degree-of-freedom trajectory simulation. Trajectories were computed for three bodies haying the following mass and aerodynamic characteristics:

II

ILI

-220-

S,, ..' ,'.

I
TiUB JONUS HdOPKINSl UNIV|IKIITY

APPLIED PHYSICS LABORATORY

*,

Characteristic C A CN

Body A 0. 104 2

Body B 0.104 0. 88

BodyC
0. 104 2
'4

C In1 q AXid S30 m

-1 0,114

-1 0.261 3

-2. 28 0, 114 3 31.1 3.5 30

31.1 3.5 300

31.1 3.5 30

d
1

Body A is the body used for all examples in this report, and additional characteristics for this body are given in Table 2-1. Bodies B and C have the same value of F. (s:ee Eq, (4. 5-62)) as has A, Therefore, the lock-in criterion (based on trim conditions) for a combined asymmetry consisting of a center of gravity offset and an aerodynamic trim (see Eq, (4, 5-61 )) is the same for all three bodies. However, for a given altitude, WA is higher for B and C than that for A by a factor of about 3, and therefore, for resonance at high altitudes, B and C tre more likely to attain trim conditions than A. The trajectory initial conditions used in this study are VE = 20 000 ft/sec, YE = "300, and hE = 400 000 feet. An in plane combined asymmetry was used for the first set of calculations, The asymmetry consisted of a center of gravity offset (r/d - 0.001) along the +Y axis and a positive yaw moment coefficient (C,,o). The altitude for first resonance was varied from about 40 000 to 140 000 feet by changing the initial roll rate. For each initiRl roll rate, C, was increased until lock-in occurred, Out of plane asymmetries were considered using the same

'

221

...

_41111111.. .. . ..

...

...

. ,.

VI Msr JOWsi MOPIN G VAU,,AII

APPLIMO PHYSICS LABOHATORY

uenter of gr.vity offset with positive and negative values of pitch moment coefficient (Cm ). Values of eo are r'Lvctly reantecd to Cn nnd C mo(see Eqs. (4, 5-67), (4. 5-(i8), rindt (4. 5-69)). ThreU aspu1:tS of trim theory are important in the

establishment of lock-in criteria. 1. Maximum angle of attack occurs at X or W5A = PE 1

2.

At X = 1 the amplification in angle of attack

S;o;or/;) is l/D.
3. At X : 1 the plane of the angle of attack has rotated 900 from its (defined by A6) value at reentry (K - 0).

The results from the six-degree-of-freedom trajectories for body A with PE =l0000/sec (corresponding to first resonance at 89 000 feet)show that none of the three assumptions was valid for this resonance altitude. The magnitude and plane of W lagged the trim values and the peak amplification was much less than l/D. The sketch shows a comparison of the K and A time histories as given by trim and simulation results. The trim 9 history reached a peak value at X - I where lock-in was assumed to occur and then decreased with increasing time, since D increased with time. The trim value of IA0I decreased from 1800 to 900 at X = 1 and remained at 900 as time increased, The simulation results show that a built up to a maximum value but lock-in did not occur, Following amax, the missile began a coning motion with no large excursions in roll rate. The peak ev was about one-third of the trim value at the initial first resonance altitude; the peak a occurred about 1. 5 seconds

222

IJIVSNI'MY? INit JOHNSiHOPKiNSi

APPLIED PHYSICS LAMORATORY


SILVIO SPRIG MANTLANO

"TRIMTHEORYTWITH
LOCK-IN AT
,, -

SIMULATION RESULTS

.90,

'-

\I II
Ii\

I
00

\\\

w00

00

TIME FROM REENTRY

later, or about 15 000 feet lower in altitude than that corresponding to first resonance altitude; the plane of rotated through an angle of 90 about 0. 7 second later, or about 7000 feet lower in altitude than that corresponding to first resonance altitude. The angle of attack amplification effect results in larger values of the asymmetries required for lock-in than those given by trim theory. Under some conditions the lag effects would be similar to lowering the first resonance altitude and, thus, asymmetries required for lock-in based on trim theory would be too high. However, for altitudes where first resonance usually occurs, the amplification effect is predominant and the required asymmetries for lock-in based on trim theory are too low.

223

I"9 JOHN$ HOPIUNSLUNIVrFIVIIY

APPLI'D PHYSICS LA1ORATORVY


*.IIg*R *9RIP MARYL6AN

The condition required for lock-in is given by Eq. (4. 5-58). If trim conditions are attained, then Eq, (4. 5-5B)
may be simplified to Eq. (4. 5-61). Both equations are plotted in Fig, 4 -17 for body A for an in plane (k = 1 ) and an out of plane (k = / 2) asymmetry. Note that the assumption that trim conditions exist implies very lnrge vahues of the roll amplification factor G at thu higher altitudes. Also shown are the at r/d values required for iock-.In obtained from simulation results, At the lower altitudes "trim theory provided the correct criterion for lock-in for in-plane asymmetry; at the higher altitudes the values of r/d required for lock-in obtained from simulation results were someowhat higher than those given by trim theory, In fact, the lock-in boundary is very nearly.a constant G contour. The results also indicate that the ao r/d value required for lock-in for the out of plane asymmetry is about twice that required for the in plane asymmetry as predicted by trim theory. Similar trajectories for bodies having in plane asymmetries were calculated for bodies B and C for an altitude of 136 000 feet only. 'These results (Figs. 4-18 and 4-19) show that bodies B and C achieve an effective roll amplification of about 18 compared to a value of about 12 for body A. Although B and C have the same response, C was much closer to a trim condition than was R. Note that for an altitude of 13U 000 teet the trim condition for C corresponds to G 27, whereas trim for B corres pends to G , 60. These results indicate that w.A is one factor which affects thu attainment of trim but it is not the only factor, and much work is required in order to define conditions for which trim theory is applicable, Trajectories were also computed for body A for a of a product of consisting asymmetry in-plane combined inertia andsa center of gravity offset. In this case Ci8. Was held constant at 0. (6360 and r was varied until lock-in occurred. The altitude for first resonance was 89 000 feet. The results show that lock-in occurred at a value of

'

-224-

S.

.... ...... .... .... ..

TH4 JONN1 MOPRIqig wNNVtRhI1T

AI'PLI[D PHYSICM LABORATORY


SiLVbe NPAING MASL6A..D

8 r/d a 2.6 x 5 If the asymmetry had been an aerodynamic asymmetry and a center of gravity offset, the simulation value of a r/d would have been 1, 2 x 10-5 (see Fig. 4-17). This ratio of about 2 to 1 between the two types of combined asymmetries is nearly the ratio obtained using trim Eqs. (4, 5-61) and (4. 5-125). Therefore, although trim was not attained for either set of asymmetries, the relative behavior in terms of lock-in criteria was properly prre( o.'d using trim equations, Until a.Iltional data become available regarding the applicability of the trim equations, the following comments appear to be valid: I. Trim equations pro,.ide a valid criterion for the lock-in condition provided the roll amplification factor implied by trim is not too high, 2. At resonance altitudes where trim conditions imply very high values of G, trim is not attained, and trim theory predicts asymmetry boundaries that are too low. 3. Values of go r/d required for lock-in at very high altitudes tend to follow a constant G contour rather than that defined by the trim equation. 4. The limits of G for which trim conditions are For attained are probably a function of many variables.

the bodies, trajectory, and initial reentry conditions used in this study, values of G greater than 10 to 20 were not attained. Until more work is done on this problem, trim theory may be used to estimate the asymmetries required for lock-In. However, the results for conditions corresponding to values of C greater than about 20 should be considered as questionable. 5. For an out of plane asymmetry, the required o r/d for lock-in is larger than that for the in plane asymmetry as predicted by trim theory. Also the relative magnitude of combined asymmetry, a and r, and combined asymmetry, o and r, are proptrly defined by trim equations. -225 -

,I

UNivalmIrf IP"I JOHkNG HOPKINN APPLIEDI PHYSICS LADORATORY

SILVI

00-OMPLN

'

!I

W..

II
22

t5

"i,

Preceding page blank


S-227 -

tHI

J014N4 H4OPi4IdS UNtIVtlMBi1V

APPLIED PHYSICS LABORATORY


SILVIA

sIpol'" MAtit&ND

'

.2

.3

.4

-3

.2 U

13

4
,

COT )I

Fig, 4.14

VARIATION OF Xe WITH D COTO.

I2
-

22B

THI

JOHN%14 HOPKINSU LNIVIRIIIT

APPLIED PHYSICb LANCRATOIRY

t I

qr4I

00

229

A4IPLIED PHYSICS LAbh.)ATOPtY

or
-4ow

0I

I
to

ILI

23

2 30 -0I
-

I ~'

APPLIED PHYSICS LABORATORY

10- 4

G---r----13-

OUT.OF.PLANI

10I

1001
10.61
-

Fig. 4-17

ROLL LOCK-IN CRITERIA, BODY A

-231.2~4

TH& j: ON

PII '

NI,

"

4P PLIE) PHYSICS ..Af3RATORY

10r

OUT.OF.PLANE TRIM ITHEORY)

02

I
4.
-,

EQ 46-SB
-

EQ 4.B-61 RESULTSI

- - -SIMUJLATION

40

60

s0

100

120

140I

ALTITUDE Ithousiands of feet)

Fig. 4-18

ROLL LOCK-IN CRITERIA, BODY B

-232-

HOPKINS UNIVSftsI1V Thi JOHNVI

APPLIED PHYSICS LABORATORY


SILVIO SPRING MARYLAPWD

-4

IN-PLANE

10 OUT.OF:PLAN.E TRIM (THEORY)

40 60

eo

10

iao

40

'

OLL LCK* Fig. .19

C CRIERIA BODY5

-233-T

SIUATO

E'4.I6

Ik

r
'm
,.Ot4NN
MIOING

UNIVUIrTv

APPLIID PHYSICO LABORATORY

IL

2351

w,......

-23

I
T~4UJOINWI M@PgMINSl w.IvmsIYElv APPLIED PHYSICS LABORATORY
SILVRR SPRtIN MARYLAND

SUMMARY

Examples illustrating the material presented in Section 4 are presented in this subsection. The following specific proibIems are considercr: Example 1 - Minimum energy trajectory for nonrotating earth Example 2 - Ballistic trajectory for rotating earth Example 3 - Body motion (oscillation) characteristlcs in the exoatmosphere Example 4 - Histories of reentry velocity and related factors Example 5 - Body loads as affected by a combined asymmetry (center of gravity offset and a distorted body) Example 6 - Roll lock-in aN affected by the combined asymmetry of Exclmple 5 Example 7 - Roll lock-in as affected by a cornbined asymmetry - center of gravity.A offset and nonzero product of inertia

'I

Precedin, page blank

-237

*1i

1NHi0MJOiS PIOPRINS UNI'VSftSITY

APPLIED PHYSICS LABORATORY


GKV111I 8P11'kQ MAqr~kANO

4, 6, 1

Example I

Compute the characteristics of a minimum energy trajectory (MET) for a nonrotating earth for a launch from
0L 0, A 1 = 0* to a target at Tho range is(Eqs. (4. 2It = 3U10 nmi,

OI

450, AI = 450. 7) and (4. 2-1 8)) 600.

The range angle is (Eq. (4, 2-18)) 280 -- 2 8e The azimuth angle at launch is (Eq. (4, 2-20)) 3540. OLI For an MET, the initial flight path angle is (Eq. (4. 1-10)) o * 450
-

150 = 30.

From Eq. (4.1-11) o 0,815 and, assuming the initial altitude to be sea level, the initial velocity is V
0

- 0. 815 (26 000) - 21 200 ft/sec. (4. 1-22))

ha

The peak altitude of the trajectory is (Eq. 625 nmi.

The velocity at peak altitude is (Eq. (4.1-24)) Va -- 15 500 ft/sec, or about 73% of the initial velocity. The time to impact is (Eq. (4.1-17)) t secondrc. 4.6,2 Example 2 1480

For a rotating earth, determine the characteristics of a trajectory fired with initial conditions V,, Y0 , and OLI (relative to earth) computed in Example 1. The inertial quantities V,

v', and

at launch are

(Eqs. (4. 2-7), (4. 2-8), and (4. 2-10)):

Preceding page blank


-

239 -

I.

Thl JOHNS HOPKINS UNIVSIRIBIT

APPLIED PHYSICS LABORATC)FtY


UlrLy11IIim q m ISe Yr.,d L.,.n

t iL
tan
tan
=

. . 1

1 5 20( l)

, 5 4 =n 1 1 4 2

21 200 cos 30 sin 35,4

1.142 tan 35.4

OL
VL

39). ILI
1520(1)1 O.036

1 + cos 30 sin 35.4


V L m 1.036 (21 200) 22 000 26 000
s
=n'

22 000 ft/sec

21 200

-0.0363

sinX sin y
y

22 000 0, 963 (0. 50)


90

0. 48 2

=2.

Therefore the effect of earth rotation is to increase the initial inertial velocity by about 3. 6%, to lower the flight path angle by 1. 10, and to increase the azimuth angle by 3. 70, These changes would have opposite signs if the vehicle had been fired in a westerly direction. Using VL and V1 for V. and 70, the equations given in Section 4. 1 apply. Howeve'r, note that the trajectory is not an MET since Eq. (4.1-11) is not satisfied. For the given VL a minimum energy trajectory would haveVL = 28. 1 so that the trajectory being considered is slightly high conmpared to an'MET,

-240-

..

.'.

APPLIED PHYSICS LABORATORY

The nonrotating earth range angle is (Eq. (4. 1-7)):


28o (7.40

The corresponding range, r0 ( 2 0 ) =4020 nml,is the arc length traveled in inertial space (or the range on a nonrotating sphericnI eiarth for initial conditions VLV, ~L The impact time is (Eqs. (4, 1-5) and (4, 1-16))

t=,1660 seconds.
The impact latitude is unaffected by earth rotation
and is
450.

The Impact longitude is (Eq. (4. 2-16)) A,

49,. 40.

The impact range is (Eqs.(4. 2-17) anti (4. 2-18))

370

nmi

A summary of the results is given below: Non. otating Quantit Flight time (seconds) Impact latitude (degrees) Impact longitude (degrees) Impact range (nmi)
-Earth

Rotating Earth 16(60 45. 0 49. 4 37B0

14B0 45. 0 45. 0 3610

rotation is to increase the range by 170 nrni and to increase the flight time by 180 seconds.

For the initial conditions used, the effect of earthI

46.3EapeThe typical reentry body, having mass characterinthe exoatmosphere, At separation, the body Is spun about its longitudinal axis at a rate of 3600/sec and,

-241-

?NIs JOQ4m

uNIVIRU11tY WOPIKINBl

APPLiEb PHYSICS LABORATORY


kIL~Ve OftNe MATLAwti

unintentionally, is given a transverse rnte of 600/sec.


Determine the characteristics of the body motion during its travel in the exoatmosphere. The only forces acting on the missile trc weight and the forces resulting from the body rates. Under the influence of these forces, the body is gyroscopically stnbilized, and the transverse rate imparted at separation results in the conical motion described in Section 4, 4.1. The period of the motion is (Eq. (4, 4-25)):

SI= 5.12 seconds.


The signal strength of a tracking radar varies with the attitude of the reentry vehicle. Therefore, the radar signal strength would oscillate with a period r = 5.12 sec-

s.

onds
The somicone angle of the space cone is (Eq, The body rates measured by, on-board 590, (4. 4-23)) 0 rate gyros have the following characteristics (assuming a symmetrical vehicle): 1. The roll rate is constant and equal to 3600/nec.

2, The pitch and yaw rates very sinusoidally about zero with peak values of 600/sec. The pitch and yaw rates are 900 out of phase. The period shown on the trtices is (Eq. (4, 4-24)): TB

1,11 seconds,

If the body were spin stabilIzed so that the total angular momentum vector It were along the flight pa-th_ direction at reentry, during the early phase of reentry a Iowever, the pitch would be constant and equal to 51). and yaw (inertial) components of a would oscillate with a half-amr-litude of 590 and n period of 5. 12 seconds.

i,,

I
tI4| .JOHNS WOPSI~i4 UNIVSPIY|I'

APP'LIED P'HYSICS LABORATflRY

'I
4. 6. 4 Ixample 4

I
I

Determine the effect of the atmosphere on the motion of the reentry body using the following reentry condi-

tions,.
hE V, -,00 000 fleit .

2(00 00 ft/se

.300
PE 0/sec and ',h01'/sec (6. 2H rad/sec)

I
, J,

0 given in Table 3-1 and Atmosphere brd Use the 1962 Stnd To anle o pt-b.m. in T.Able, given the mass Lind aerodynamicv characteristics

conical motion rs problem. not in a The 0, the thes vehicle 3Io was SincemTb as described in previous angle of atta ck tv remains constant and in the original plane a pitch nd yaw angles of attack remain constant) (Inertial bw until terodynamic forces begin to affect th e motion byI as the beginning Using tu0.d99 th in dtcb y cause of then 01.a decay, for . PE 3600/sec: - (Eq. . 64.5)) -5 21 x 23. f0e7

1(

Po

From Tfable 3-1, this density ratio corresponds to an altitude between 300 000 and 350 000 feet. A more complete table shows the altitude to be h = 332 000 feet. If the Initital roll rate were zcro, Eq. (4. 4.-41 ) could be used to obtain the altitude at which the decay begins.:

' '

0. 02 (.30o) (32. 2) 1(o.25)

. 5(2)(0o 4) (21T-Y-

2 96x1 -:3 e t

)xo

fe

'

I
-_243_-

------

I.

~APP'LIED PHYSICS

HOPKINS J1NI~IPofll THI .JOHN&I

I.AEpRATCnRY

Using thle charancteri1stics of' thu 1 !)6i2 Standarid Atmos~phere, this value or 11 corresponids to nn nititudu of' about

30/ soc spin 350 000 feet. rkherufrc, the uffuct of' tih:o~ ra~te Was to de'lay th0 W deOCay by ab-OUt 18i 000 fQLt.
VIW C( )1IflptCd 0811Wj Thu ;; lu(Ca Ili qtotr iL'LA 11. \' Eqs. (4. 4-44) and (4. 4-53), In UidUl' to evaluateW thcsuL

equations, thle variations of w

rand K with alltitude are

Those characterlst&H, obtnined using Eq~s. (4, 3-6) and (4, 3-23) and the atmosphore defined In Table 3-1 , a-re listed in Table 4-1. Prvom Eq. (4. 4-45) and Table 2-1, tile danmping constant is: required.

About 60% o f C:i comnes from tilt CNr termi ind 40% from
the Cm,
q

term.

Trle W dectly history for tile rolling body

was computed using Eq. (4. 4-5`), 111d flt)u VUSUlS arc shown in F ig. 4-20, Also shown is tile ease of the nonlrolling missile ond the ctise of n rolling miissilec with ing term is negligible for altitudes above about 150 000 feet. Vor nltitudos bolow about (GO 000 feet, the W history for tht. rolling ca so (with damp ing) is noarHy thle sarnc 11s that for the nonrolling case. 1However, nt altitudes above 60 000 feet, thle angle of attack at a given altitude for the rolling body is significantly !ifgher than thnt. for the nonrolling body. This fact could he inmp ortaint for' some( nmissilos in which the punk lateral lond occurs at these altItudes. For the body studied, thle peak lateral lond resultK 4 (7.6 + 0,75) .0 09)0
-

From Table 4-1, this value of K correspondH to ain alti tude of about BBi 000 feet, The altitude for manximuni is the same whether or not the missile is rolling, Howfor thle over, as ra result of the ev histories, g1
-

~.tt

244

I
?MR JOHI MOPKNINNUNI|NIrTv

APPLIED PHYSiCS LAUORATORY


SLsim i Mml ,..,,,Lan

I
rolling missile is about 45% grenter than that for n nonrolling vehicle (Eqs, (4, 4-60) and (4. 4-63)). lor = 101, 1E g~n for a rolling vehi.cle is 2. 75g compared to 1, 9g for a nonrolling vehicle, The reentry velocity, Mach number, and Reynolds number arc important environelental factors that affect hody Ion Its and tciperrturu, The time hi story of these characteristics estimated using Eqs, (4, 3-5), (4, 3-7), and (4, 3-10) and the atmospheric characteristics given in Table 3-1 are listed in Table 4-1, The Mach ni-'! r history shows that the vehikle Is in the hypersonic rrlime (M t (6) down to an altitude of about 15 000 feut aid is supersonic fronm 15 000 feet to impact. The Reynolds number history shows that transition on the body woulcj be expected at an altitude of about 110 000 feet (RL ft 10 ) with lamknar flow at higher altitudes -ind turbulent flow at lower altitudes. The heating rate to the conical surface of the body would increase by a factor of 2 to 3 when the flow becomes turbulent, and therefore asymmetries that result from heating would be expected to increase rapidly at the transition altitude. The aerodynamic frequency data show that roll resonance

x~~ 1i)

occurs at nn altitude of about

1 35 000 feet. It should be noted that the vehicle is approaching a second resonance condition at tImpact, and any asymmetries that exist near altitudes of 135 000 feet and sea level tare likely to be t.implified. The computed values of M/h. show that free molecular flow regime begins at altitudes well above 400 000 feet (M/Reo 10), transition flow exists down to an altitude of about 250 000 feet (M/'"e `, 0, 1), slip flow exists at altitudes from 250 000 to 140 000 feet (M/V'-.' 0, 01), and continuum flow exists at altitudes fromn 140 000 feet to son level, The time from 400 000 feet to impact is about 49 seconds compared with about 39 seconds if the -itmosphere were absent. Therefore, tne atniosphere int vcased the flight time by about 10 seconds,

:1

2-45-

N1.

MOHN.

HOPKIN.

UNIVEISIT

AP'LI[D PHYSICS. LAIORATORY

Although not shown in the tahic, severni other importnnt consi deratiovnts may 130 obtninud from the equa tions given In Section 4, 3. Using the rntio qs/qtlonx
0,1 as Lin Indication of eondlitiOnH wherLe heattinog is s ignif'icant, the n1titudes ore computed (PI i . (4,. 3-11) anl(. (4. 3-10()) to be about 190 000 feet for the ueginning of thw hunting peril ld 1 1l)Olt t111 1 3 000 f[U t forV If . end of heatoing, With the n11;dm11u.| hujntink Vntte ot M 1 nhltitudtI or :Ibout H11 000 fQuet, Thi, h0logitu lt innl I .et'nt ( f s (4, and (4, :3-14)) Is maximum at an altitude of Tbout :5 000 feet where m -54g. The inflection points in the gN versus time curve (Eq. (4.,3-19)) occur nit altitudes of 5B 000 antid 19 000 feet. The results obtained in this problem nie usunim. marized In Vi'g, 4-21. Although the nltttude,bankds shown . .

vary slightly depending upon vehicel

tory diharcteristics, the results are typienal of' high J3 re-': en~try bodies. Note thant practtectlly all of thu quantitttem of interest occur at altitudes holew about 1,110 000 f'eet and

ge-om-etry mind trajoe-

therefore atre in the continuum flow regime,

Both laminar

and turbulent flow conditions exist In this altitude region, SaUl three Mach number regimes occur (a'though only two occurred for this trajectory), and both oscillatory and trim types of body motion may be important.

4. ,5 .
kI

hxn m l c

Suppose the reentry vehicle has a center of gravity offset from the body centerline of 0, 042 inch. lF'urthermore, assume that as a result of exontmospheric binst effects the body is distorted so that the body centerline is an arc of a circle and that the tangent to the centerline changes by 0. 250 from base to nose tip, Use the saime

initial conditions as were uoed for Example 4 except use


PF = l0000/sec. Assess the probability lhnt the missile 'or this problem w-il encounter excessive lateral loads, excessive. as 20g 9L define we will

-246-

I
I 1"~HkJCOHNI MOOK~iN5 UNIV9111ITV

II

APPLIED PHYSICG LABORATORY *'LY111 P S9NG MAfYL.A~b

From Eq, (2-35)'the Um

restulting from tile bent

body is 0. 002. We will assume that the lateral location of the center of pressure and center of gravity are unaffected by the bent body. Flrom Eq. (4. 5-65) the trim angle of attack resulting from CM iq: 0 0.0,02 57 3 0 0M 2x 0, 114 x57,0,5,

S18

The trim angle resulting from the center of gravity offset (4,.5-64)): (Eq.

e o

104 0.042 4 =I0, 42 x 5 7,'3 = 0.0 261 2 x 0. 114

Therefore, the total trim angle Wo is very nearly the angle resulting from the bent body and we will use 0 = 0, 50. The equation for gl is: CN S

4L
For the moment assume that no amplification in a. oceur.c, so that A = 1. For this case, the largest possible vtlue 'mx cI: I From Eq. (4.3-13) occurs at q of at P/Po

2116 2110

(.)

= 0.

237, corresponding tW sn :,!tit ide


Tke iaanilitu',:

of 35 000 feet and a temperature of 394 0 R. of q is:


-

q max

1000 (20 000) (0.5) 9330 (394)

54 400 fl/ift

and g1, g- max :9. 2g

-247-

,1,

THR JOHNS HOPKINS UNIVERSIT

APPLIED PHYSICS LABORATORY

Excessive values of gL will not be obtnined unless values of A in excess of 2 are obtained, Amplifications of this magnitude will be encountered if large excursions in roll rate occur. The possibilitios that must be consiclered arc: 1.
2.

Roll lock-in at first resonance,


Rloll I )ck-in -it second reso( un ce,

3. 4.

Spin up with lock-in at some intermediate altitude, Spin down with lock-in at a negative voli rat,.

I
,

Assuming that lock-in Is attained, trom .1Eq, 1/iD and D may be evalur-':,d usihig El, (4. 5-16) A (4. 5-60). Substituting these equations and th c utivtlo ' 1, excecssiv, for q (Eq. (4, 3-8)) into the equation 1. loads will occur when the following inc.untilty is sattified, 57,3 x 20 (k0o; 2 C gd, A" d

C NIN N7 E
where co is expr.isad in Jectory p ar:metter ., feet:

1 '

,y

,d.oree...
LI.k,

Substituting

'orthe' U 1.

,:I , rmt'id.l:lan and as,iur',nj! tl

),, ': : .' OHL;.:,t

1..ll ,

S0. 04

I
leIF;

This Inecqualit, is plotted in Fig. -1-22. It i..i u,; ,. .', 0, 50 is large, (h:i,;U-!1 to ,'sult ill eXC"-;:V' I.i.... that io 1c at some altitudes, Tho fir':8t od second rescmntc( .IIi are the altitu,les for whiclh WA =-,

PI

Ise..

248

M 11-1 "--"-NA

"APPLiED PHYSICS

7M` JUNIN

MOPtINS

WNtV9RttTY

LABORAT;-)RY

[':

Values of wA were computed using Eq. (4. 3-23) and the results nre shown in Fig. 4-23. First resonance occurs -at a. altitude of about 89 000 feet, and second resonance occurs at an altitude of ahout 4000 feet. Therefore roll lock-in at fir'st resonance would result in excessive loads, hut lack-in it second resonance would nlOt. i['Depending upon the value of Oo, the roll rate may

istics. However, the magnitude of Ap in this case will be underestimated slightly since the amplification in o and the change in the angle 0 are neglected. Using Eq. (4. 5-93) and assuming values of 900 and 2700 (which give the largest values of Ap) vL-,iues of Pr were computed and are shown in Fig. 4-23. From this consideration it is found that the asymmetries are not large enough to result in roll reversal. Therefore, resonance at negative roll rates may be eliminated as a possible condition that would result in excessive loads. However, if the vehicle does not lock in at first resonance but spins up, resonance may occur at an altitude of about 13 000 feet. From Fig. 4-22, resonance at this altitude would result In excessive loads. Therefore, the problem is reduced to an evaluation of whether or not lock-in will occur at altitudes of 89 000 and 1 3 000 feet. The criterion for lock-in is (Eq. (4. 5-61)):

0rd

(I -KR)

Using the data given in Tables 2-1 and 3-1 and the trajectory char cteristics, the value of F 1 (Eq. (4. 5-62)) is 5.4 x 10- for h = 89 000 feet (H = 21 700 feet) and is 4.7 x 10-6 for h = 13 000 feet (71 25 000 feet). Using r/d 0. 042/12(3. 5) 0. 001 and expressing 0 o in degrees:

-249-

S.

I
mml mmi m il l ! l

...
ml~~ l ii m

. . .. . . .. . ., ,.
d| i iI[ii INi I III ! I'a]

THI

JOHN$ HOPKINS UNVIISITY

APPLIED PHYSICS LABORATR)

S(1 0. 3 1

Kl
f'o h1 W) 000 foet

(i
> 0, 27

K H)
k
for 11 1 :R000 1'cct,

Using the data of T:ible 4-1 a plot of K versus altitude shows that KR 0, 077 at 8,9 000 feet nnd K 0 2, 5B at 13 000 feet, Therefore, the criteria for lock-in are:
-

S-

ok

0, 286

lfor hii s19 000 feet


for 1 = 13 000 feet.

S-0.427

j
I
.IZ

The value of k depends upon the value of 00 and, for lock-in to occur, k must be po-itLve for I<R <1 and negative for Ki > 1. For this prohleL we must assume that 0o may have any value. Therefore, we must consider the following possibilities: Resonant Altitude 69 000 feet 0o -00 !}00 2701 13 000 feet 2700 k 1 0, 5 0, 5 -0, 5 -o 0, 286 0, 572 2 0, 572 0. 854 ,

The other twu values of 00 having negative values for k for the `13 000 foot case are ruled out since the 00 = 2700 value is required to provi-to spin up during reentry (see Eq. (4, 5-;)3)). O'f the four possibilities onl~y the==0o :00 case with resonance at 89 000 feet satisfies the & crite rion, In the preceding discussion w' considered only inplane and out-of-plane conditions, ].or intermediate values

I I
r

-250-

3
,. .. . ', , '. ! " " .;', . . .. .. ' p .. '.

S.

WNIVIRIIIII1TY THOt JOHNSI "OPKIP411

APPLIED PHYSICS LABORATORY


SILVIOt 6PINIOtm MANTLANn

IJ

of 00, io may be divided into in-plane and out-of-plane components. Any value of 00 for which the component of ;o in the 00 = 01 direction is z 0. 286 will result in excessive values of gL- The possibilities are shown in Fig. 4-24. Any value of' 0 defined by points from A to B 0 (shorter arc length) will result in excessive values of gLSince any vnlue of 0o is equally likely to occur, the probability of encountering excessive 's: iL . 283 2 cos"1 0 360

F
f

0.3

4. 6. 6

Example 6

Sr :!igiven '["

For the trajectory used in Example 4, estimate as a function of PE the maximum allowable &0 combined with m 0. 042) inch so that neither lock-Ln at first resonance nor spin through zero occurs. In view of the discussion In Section 4. 5. 10, assume that the criterion for lock-in is the trim equation (Eq. (4. 5-61)) or G -- 20, whichever gives the larger value of &o. The critical conditions to be investigated are the 0o S 00 case for lock-in and the 00 - 90" case for spin through zero. For the lock-in case, assume that H = constant 21 700 feet. Substituting the mass, aerodynamic, and trajectory terms into Eq. (4. 5-58), the criterion for lock-in becomes:

2.65 (1 -KR)

If trim conditions are attained, the criterion for lock-in is (see Example 5): P4 0. 31 (1 KR)2-51

'I
='"" ................................................ 4_ _''''..... '...... "'.

Tlti JOHNSHOPK(INSUrRII

APPLI1EDl PHS SS

LAIBORATORY

20 in the first eq~ti-fl Sett IngG0 the ssme solution to ii0 2.3 0. 3

th.

wo

qaifSg

A~

or

70 000 feet (Sce corresponding to an altituido of about > lower than 70 000 feet (or K 4 ~an Table 4-i). At ,11titudies is higher Rltitiide8 at used, 0. 18~3) the trim equation aqllatiofl is used with 0 20. 70 000 feet tile noe1trity correponling.ThLiF~or each xesonant altitud-o the equation:~ the frorn tial roll raite ME'Y bno Computed WA

Ix
c Thle valuesore where WA Is given in Table 4-1. i. functio Fi. qurdfrlck-in are shown a~s 4-25. from asymmfeThecangein roll rate resulting Spin through nd r/d is given by Eq. (4, 5-133). &p is negative tries; 0 Fi if reentry zero roll rate will occur during condition restr'ictive most Using thle &p : pE PE In units. and expressinlg for (00 9011 and K rK )andres~pectively, critevion the of degrees and dog sec, spin through zero roll. rate is: -41 spin through zeru r'oll rate Tihe values ("f i; required for are shown in Pig. 4-25.

4and

THIC JOHNS HOOlKIN LNIV9I"lTV

APPLIrU PHYSICS LABORATORY


*IIgft RPAIN6
MAUAND

This problem illustrates some of the trade-offs which must be considered by the designer in selecting a roll rate for the reentry body. If the roll rate is very low, first resonance occurs at a very high altitude, where large values of 00 may be tolerated without encountering lock-in. However, vehicles with low values of PE are susceptible to L-in through zero roll rate. On the other hand, for moderately high values of PE, large values of 0 may be tolerated without spinning through zero roll rate, but oo must be very low if lock-in is to be avoided, For this particular problem the best roll rate Is PE = 7600/sec where the missile could tolerate up to 0. 60 in &o without encountering either adverse condition, A second possibility would be to select a value of PE much greater than 22400/sec. In this case spin through zero roll rate would be unlikely, and lock-in is avoided by using a roll rate that Is much greater than the maximum aerodynamic frequency. This technique is sometimes referred to as "overspin." 4.6. 7 Example 7

i1 '

Using the trajectory of Example 4 with pE = 10000/sec, first resonance occurs at an altitude of 89 000 feet (KR - 0. 077). Determine the values of 8 and r for which lock-in at first resonance may occur. Assume that the trim equation (Eq. (4. 5-125)) is valid. Compare the results with o and r given by Eq. (4.5-61). 5.4 x 10-6 (H = 21 700 feet). From Eq. (4. 5-62) F 1 Using the mass and aerodynamic characteristics &iven in For 8 in . 26 x10-". 4. 87 and F 4 Table 2-1, F 3 degree units, Eq. (4. 5-125) becomes:

rI
ei 279~.l 1-2,5
The ea boundary is shown in Fig. 4-26. At a given value to- must be greater thanl6min but less than of r, Lock-in cannot occur for any value of -253If r < 0, 06 inch. i8

.max.

I,,

.j..

THIr JOHNS HOPKINI UNIMiftalV

APPLIEn PHYSICS LABORATORY


S~kIRVSSPOIfNU MARYLANDI

By compnrison the trim equation for comes:


-

o and
0

be-

0.28(0 x10-3 r/d

The oo boundary is also shown. It Is noted that &0 Lhas no Omax boundary, and the W. boundary is less than *6min Single points on each boundary obtained from sixdegree-of-freedom simulation resulas (see Section 4, 5. 10) are also shown. The (G. r) point was obtained by holding r constant and increasing no until lock-in occurred. The (;6, r) point was obtained by holding q constant and varyIng r until lock-in occurred. Note that the real boundaries are somewhat higher than the boundaries based on the trim equations as discussed in Section 4.5.10. For other values ofEpE, the ea boundary must remain within the x and &m asymptotes shown in Fig. 4-26, and rmin is shifted right or left depending upon the value of KR. Using data given in Table 4-1 and the relationship between pE and WA at resonance,
WA

E,

rmin is shown in Fig. 4-27 as a function

X
of P' For r
=

0, 042 inch used in the previous problem,

lock-in could exist only for values of PE that result in resonance very near to the altitude of maximum dynamic pressure (35 000 feet). For current reentry vehicles, values of pE are selected such that first resonance occurs at altitudes above 100 000 feet. In this altitude range rmin is insensitive to PEI and the &6 boundary shown in Fig. 4-26 is applicable to first resonance lock-in for all values of pr!,

-254-

[
u n n mI I

I
TH
JOINNI OPImNI, UNIVRRSIT
BLS1U SPPNIIi MAImAN,

APPLIED PHYSICS LABORATORY

less than about 10000/sec, For Px 1000*/sec considered In this problem, second resonance occurs at an altitude of about 4000 feet (n'nglecting a change in roll rate during reentry) wl-ev'e rmin is nearly a factor of 2 greater than rmin at fir.7t ;"4sonance.

4 I
1

,1
I

255!

2-55

!
* *.-I ~ ;

.4ui~q .. . ..

iMt JOMHS 'OPK IPN4UIvItltT

APPLItD PHYSICS LiUOPAIO Y


SIL.V, SpRING4SMAIVLAND

REFERENCES

4-1

L. S. Glover, "Approximate 1ce-Entry Velocity and Heating Equations, " J. Spacecraft and Rockets, January 1966, D. T. Greenwood, "Flight Mechanics of Space and Re-Entry Vehicles, " Michigan University Engineering Summer Conferences, 1964, H. J. Allen, Motion of a Ballistic Missile Angularly Misaligned with the Flight Path upon Entering the Atmosphere and Its Effect upon Aerfdynamic HeatIng, Aerodynamic Loads, and Miss Distance, NACA TN-404B, 1965. R. B, Powell and R. L. Smith, "Dynamics of Spinning fe-Entry Bodies, " AIAA Preprint No. 64470, 1964, H. I. Leon, "Angle of Attack Convergence of a Spinning Missile Descending Through the Atmosphere," J. Aerospace Sei. , August 1958. L. S. Glover, Analytical Expressions for the Effect on Roll Rate of Mass and Aerodynamic Asymmetries for Ballistic-TypeBodies, APL/ JHU TG 560, March 1964. J. D. Nicolaldes, On the Free Flight Motion of Missiles Having Slight ConfigurationaL sy'mmetries, Ballistic Research Laboratories, Aberdeen Proving Ground, Report 858, June 1953. R. L. Nelson, Measurement of Aerodynamic Characteriatics of Re-Entry Corifigurations in Free Flight at Hypersonic and Near Orbital Speeds, Lockheed Missiles and Space Division, TR 6-90-61-37, July 1961.
-

4-2

4-3

4-4

4-5

4-6

4-7

4-8

Preceding page blank

257

.ij

THR JOHN% INl OPOINII UNIVeRIIII Y

APPLIID PHYSICS LASORATOMY

REFERENCES (cont'd)

4-9

A. E. Hodapp, Jr. and E. L. Clark, Jr., "The Effects of Products of Inertia on the Ru .1 3Behavior of Ballistic Re-Entry Vehicles," AIAA Paper No, 70-204, January 1970, L. S, Glover, "Effects on Roll Rate of Mass and Aerodynamic Asymmetries for BallistIc Re-Entry Bodies, " J. Spacecraft and Rockets, April 1965, K. A. Ehricke, Principles of Guided Missile Design, Space Flight, Vol. Ii, "Dynamc D Van Nostrand Company, Inc., 1962,

4-10

4-11

2I

$'

- 258-

Li

.... ,, . . .! . ' , .1

I"

UNtVINuITY ?MR JOHNS MOPKINS APPLIED PHY1ICS LABORATORYj


*LVIE| PIPIN4. MARYLANO

ji1
/
/

I:
--

l fl7/""uDM'

ROLLING,

DAMPED

it

00

'2

280

320

360

F~g. 4.20

-CHARACTERISTICS a DECAY

I/

259

ii,

TNC JOHNS HOPKINS UNIVaImIIe APPLIED PHYG1CS LANORATORY


SItLVI SPOIIN6 MAVIILAND

I
I
>.

400

\L)

:360

.I 320

....

INiTIAL ,i DECAY

2(10
280
-

S240

200

S HEATING BEGINS

C AERO DAMPING BiGINS

120

-,

\)OUNUARY TRANSITION

LAYEi

0go U.

PEAKIfL DUE TO _ PEAK HEATING

RATE Iq51I

40

PE A

,IT, AN DWA

10

20

30

40

60

r ROM 400 000 feit ALITUDJE (seconds)

Fi,4.21

OIGNIFICANT EVENT,

260

S... ." It. ", '' V ,...,*,,l', . ,. . . iI, . ....-Mr .. .


S,
., ,~ , . ,,,.'.:

~ ~~:?

.( ,

7Mg JOHNS"OPKINs UNIvas*iry

APPLIIO PHYSIICS LABORtATOR~Y

It

26

APPL.IED PH4YSICS LAMOR~A,'OR

SONVNOSIW 0140039

3-Y

' 'Ii (V0916J5p

26

I_
T111 JOHNG OPKINI NIE1VIRS1

mmi

04A,LILD FHY3,16 LABORA!'RY


IIL0+1PI 001014cI MAIVLIAND

!I
--

LOCI'.IN CRITEHIA

0.6

III
0.4 A

001
II
0 04OB

F0.

42.

C- DEG REES

0.21

o.B

II
0.4

i0.6

TWE JOIHN

HOPKNIN UNIVII II1"

APPLIED PHYSICS LABORATORY


tI.. 1S BPAIN* MAIIIIIAND

200 000 4

-1

'tl 2S2

L~flt(IN RES1ULTS

THROUGH ZERO RESULTS

oSPIN

150000

400

800

1200

1600

p000

2,400

2800

PE (degrees/seood)

Fig.4.25

F,

REQUIRED FOR LOCK-IN AND SPIN THROUGH ZERO ROLL RATE I2

- 264-

.i1

.................

,,,

I It4 JOH-NS MOORINGI UNIV|RAWITY

APPLIED PHYSICS LABORATORY 6L16e0 spolINO MOANnAwN

x olo

I.

~z

LLLL

IA.g

00

""

10

Q. ,

..

0,

pnoeel~p) 9V NV0

SI-

265-

"

tklHt JOMk6 ,4OKIIKNS UNIVK[R|It

APPLIED PHYSICS LABORATORY


S I 4 MAIhLAND S PRIM.VUA

0,12

ALTITUDE F4USONANCE S L

RE SONANCE &ECOND 1010(10o FEET /

"
0,08
-

! pIRST RESONANCE

o o

goo~ ro

oo

004 0,04-

00 2009080070

400

800

1200
P dlgrS.I~tSi~Ord1

1600

2000

2400

Fig. 4-27

ROLL RATE VARIATION OF rmin WITH INITIAL

-266-

Ilk

THK jONNO I4op"411

LJNIVINEIlY

APLIIID PH4YSICS LABORATORY


MANY6Ak0 SPOIN01 BSLVID

[4J

in

mi

C,

04 N

c N dl

I-,

U1W ell Mc,

C NC

i N

0,

00

0-

tM~

to

t-.

Ni

In

C 00

~
nD
C M

0000

~
mi
C"

C;

14J

cn0

0
CI

in~ 0
.

00

t-

to

C"',~

-267-

4A.

APPLIED PHYSICS LASOMATOIM


ggLVIII WRING. MANYIAND

r,.

. DISPERSION

[
Preceding page blank

3!'

?rek Jo,l

4OPKINS

WNIVII:".ty

APPLIED PHYSICS LABORATORY


91LVlIO SPAIN4 MAPYL1AN

SYMB3OLS FOR SECTION 5 Typical Units pounds pounds

Symbol A AR CA CD c. g. CL CL .C o

Definition axial force component of A in dowrnvange direction axial force coefficient = A/qS drag coefficient = D/qS center-of-gravity location lift coefficient = L/qS dCL/d& asymmetric rolling moment o/qSd coefficient A moment coefficient about body Y 0 (see Section 2.4) axis at
Cm 2 nC

inches per radian

"Cm 0
*Cm
0

I
=

CN

normal force coefficient

N/qS per radian

SdC N/d&
C no c. p. D d E e FW F moment coefficient about body Z axis at " 0 (see Section 2, 4) center-of-pressure location drag, aerodynamic force parallel to V vehicle reference length factor defined by Eq. (5-21) Napierian base weighting factor for wind Weighting factor for density variation

inches pounds feet feet

L
' . . .. e.
L

P'eceding page blank


f'P!

271

..
i

. . . ..
. wl

" - , .' " ' : ' .. .. ' ,<. Ji.

HOPKINS UNIVIRIITY 1TI4 JOHOIN*,

APPLIED PHYSICS LABORATORY

Symbol
g 9gY g g Y gZ

Definition
acceleration of gravity

Ty_.cul Units
I't/sec 2

1 2

+ gJ

21

1/2
g

acceleration along body axes X, Y, and Z g feet feet

H h 15 I

scale height RT/g altitude functions of KS L.

IX

moment oa

inertia about body axis X

slug-it.

index number P K L

sin Y
lift, aerodynamic force normal to V asymmetric rolling moment pounds ft-lb

normal force, component of aerodynamic force normal to the body centerline

pounds

0l'
ambient pressure vehicle roll rate

or3
lb/ft2 rad/see lb/ft nml 2 1716 ft 2 01
- oi*

number of altitude bands P p q


R

2 dynamic pressure, pV /2

range or gas cunstant

/0ec

radius of the earth

feet

y0

-272-

TMtr JCMNg MWPKINS UNIVCMgIVt

APPLIED PYkICS LABORATORY BGILVE. 41PRl,. MANYLANI.

Symbol
r S S. L.

Definition
rodial c. g. offset reference area sea level

Typical Units
inches ft2

IT

ambient temperature
vehile mV velocity

O
ft/sec it/see ft/se . ft / se c ;

i}I..iVR w VA/M

drift velocity resulting from winds windspeed velocity of the air relative to the missile

W X o S

vehicle weight
distance parallel to the body centerline from the nose to the c. g, distance parallel to the body centerline from the nose to the c. p.

pounds
inches

"Cg.

inches inches

'

'

lAX

X c~p. - X e.g. angle between the body :enterline and vI angle between the body centerline "andthe reentry flight path direction ballistic coefficient, W/CAS

radians radians lb/ft2 degrees degrees feet degrees

y AY

flight path angle, negative for a reentry body change in flight path angle dispersion measured along the earth's surface refers to an increment except as noted angle defintd in Section 5, 4. 2 - 2731-

S8
A 0 1

T14

JOHNt HOCPHNi UNIVEMNITY SIlV~i SliprllnI MAUlVLANl

APPLIKD PHYSICS LABORATORY

Symbol go
,0

Definition

Typical Units degrees


slug/it3

:c.

angle defined in Section 5. 4. 2


ambient density

Subscripts!:i c E o R S. L, w refers to cross-track component refers to conditions at reentry refers to conditions at zero roll rate except as noted refers to along-track component sea level refers to quantity resulting from wind effects refers to quantity resulting from density effect A dot over a symbol refers to a first derivative with respect to time. An arrow over a symbol means a vector quantity,

.1

274

'I

. .

.'.I': i

m .\

. . ..

'"

"

U
tHlE JOi~NNO NOPKINlS UNIVINSIlIt

APPLi9D PHYGICS LABORATORY

SUMMARY AND CONCLUSIONS

The factors most likely to cause impact dispersion


are Identified, and simple equations are given for estimating the magnitude of each dispersion, These factors Include atmospheric wind, deviations from a standard atmo-

spheric density profile, errors in mass, and aerodynamic asymmetries, The following examples are given to illustrate this discussion, Example I - Dispersion resulting from wind and deviation in density at low and moderate altitudes, Example 2 - Dispersion resulting from wind and deviation in density at high altitude, Example 3
-

Dispersion resulting from a trim angle of attack for a missile rolling at a constant roll rate. Dispersion resulting from roll lock-in at low altitude,

Example 4

Example 5 - Dispersion resulting from a combined asymmetry for which spin through zero roll rate occurs, The major conclusions are: th Winds and departures from the assumed density are the chie; Nources of impact dispersion resulting from meteorological characteristics at the impact site. This dispersion increases rapidly as the flight path becomes shallow, as the missile velocity is decreased, and as the ballistic coefficient of the missile is decreased. The density effect is more sensitive to these missile and trajectory parameters than is the wind effect, Equations are presented for evaluating the dispersion for any wind and density profile,

I
-:": . . . ..
!!

."
,, ....

. ..

. 7.

' ',
.... ..

""
.

' ' " . : i " .:' '' -" ..

'5

?"VM J M4

O~lll LNI 40PIN|k

BtYtl~?

APPLIE10 lHYSIIC6 LAV'IORATORY

2. The equations derived for evaluating the dispersion resulting from deviations In density may also be used to evaluate the dispersion resulting from perturbations ih missile weight arnd drag, 3, The dispersion resulting from a vehicle that encounterm roll resonance (but remains intact) in likely to be small, On the other hand, the dispersion resulting that results asymmetry a likely combined havingIs vehicle a rate from in roll reversal to be extremely large.

27

I.

II

-27-.

"t.4'JOHNS
BILVI

mOPRIN3I UNIJ 011v

SPUN.NMAEYTLAP4

Aiming instructions are required in order to deliver accurately a ballistic vehicle to a predetermined impact point. To obtain launch order s, a mathematical model containing all factors that influence the trajectory is constructed. However, even with the most elaborate trajectory simulation that can be devised, the model is not perfect and the vehicle departs from the intended trajectory, The magnitude of the miss distance from the intended impact point will be referred to as the impact dispersion. In Section 4. 1 some of the important factors which affect dispersion and which occur at the end of the powered phase were discussed (evror in terminal velocity and flight path angle). In this section we will discuss only those perturbations that occur during reentry. Accurate evaluation of dispersion requires use of a three- or a six-degree-of-freedom simulation. Simpler solutions, although lacking in accuracy, make it possible to isolate the important factors and provide insight into the general behavior of the various types of disturbances. These simpler solutions wfll be discussed here, 5. 1 Meteorological Characteristics The meteorological characteri'.ti.s that may produce dispersion are: 1. 2. 3. Errors in wind speed and direction, Errors in atmospheric density, Errors in atmospheric temperature,

Temperature is included since, in addition to its effect on density, it affects also the speed of sound and, thus, the aerodynamic coefficients that depend upon Mach number. However, the effect of the speed of sound is usually small (see Section 3. 1) and only the wind and density effects will be considered further.

t{

- 277

.. . . -. .. . ... ...

F-.

T'i9 J014NO HOPKINS UNIVKMRITY

APPLI)ED PHYSICS LABORATO'RY


S|Lk~A UPSINO MhU~L.NL'

5. 1.1

Effect of Wind Atmospheric winds are a major contributor to the

dispersion of uncontrolled reentry vehicles. A simplified analysis is given in Ref, 5-1, The important portions of this theory are summarized below. We make the following assumptions: 1. 2. The effect of gravity is negligible. The earth Is nonrotating and flat.

3,
4, 5. 6.

The flight path angle is a constant


The angle of attack Is zero.

The wind is a horizontal head or tail wind (except as indicated), The reentry vehicle is aerodynamically stable (the center of pressure is aft of the center of gravity), The weight and CA are constant,

7,

With these assumptions the flight geometry in the absence

of wind is shown in sketch (a).

A)

C~.HOR

IZO NT.AL

-JE IS NEGATIVE

(27
-

278

S. .

....

... ------

-;

=-:

-- ,-= =

me

Il

/i

In

gp,

rr

II j
til JOHNS HOPKINS UNIVM9IY CiLAIMnSP~iNQ MAMVIL~tND

APPLIED PHYSICS LABORATORY

I
The velocity vector is along the body centerline. The only force acting is the axial force, A, and it also acts along the body centerline (by definition), Suppose now the body encounters a tail wind. The body forces that act immediately are shown in sketch (b). The missile velocity

SA

NN

'V'.

awSHOWN , POSITIVE IS

(,

aii
7'W

1b)

vector and A are still along the missile centerline. However the velocity vector that produces aerodynamic forces This velocity vector on the missile is Vw"acts at an angle, /- to hebody centerline and causes a normal force, N, which acts at the vehicle center of pressure. This force causes the body to pitch (weathercock) until the body centerline and VA/M are aligned. The conditions that exist when the weathercock motion has damped to negligible magnitude are shown in sketch (c).

I.2 1

-279 -

tHE JOHNS NaPKINS LNIVnERlIYY

APPTLIED PHYSIC5 LABORATORY

HORIZONTAL

4E

E
Vw

The normal force is now zero and A is along the body centerline. At all later times the body centerline remains aligned with / however, since A acts at a slight angle to I the missile acquires a drift velocity relative to the no-wind trajectory. The force and velocity diagram including the drift velocity is shown in sketch (d). We will assume that the geometry shown in sketch (d) is representative of the reentry trajectory. At any given time (or altitude) the nissile velocity is assumed to be the vector sum of the no-wind velocity for that time plus the drift velocity V R' Furthermore, it is assumed that VR <<V. Using these assumptions it is shown in Ref. 5-1 that tie dispersion resulting from winds is given by the equation: N

V~siny

13 I

VWFw

(5-1)

280

AMF A

aI

HORIZONTAL

v A/I
vi

-vw

K
281

APPLIED PHY4'SICS LABORATOR~Y


BILVIR IAPR~IO M&AILAND

where

t iven in Trir:lc 5 -1 l

ti sgivun in Tnbic 5-2, The index number~ j refers to a particular aititudu band. To account for a variable wind with altitude, the reentry altittude is divided Into N bands where the bandwidt', is small enough that Vw may be considered constont In each band,

1<

HOR i oP4rAL P IS A NEGATIVE NUMBER

RANGE

The FA term Is a weighting factor that depends upon the altitude, altitude bandwidth, and KS., L. -Since KS, 1_.

~sinV
L. an

~P

'Is

a function of the atmosphere consideo-red

trajectory characteristic ()'E)-

minor variation), the body characteristic

For a given body

()

-282

I
I t OHI I 0t00"III uNIVI-'*1Y #4PLICrDPHYGICS StL1R SPIAOI~

LADORkTC'RY
MAA1LAPlh

a given altitude Fw for of atmos!!here, and type design and altitude bandwidth is a function of YE only. For most applications this variation may be neglected. When this is done the values of Fw should be evaluated at the lowest value of TEt used In the missile design. The dispersion Therefore, aldecreases rapidly with increasingdesi. though the percentage errore in 6 may be lrge when this procedure is followed, the magnitude of the error in 8 will be small. If the altitude bandwidth is constant, the magnitude of Fw it indicative of the sensitivity of 8w to winds in a given altitude barid. F'rom Tcble 5-2 it is observed that the most sensitive band corresponds to the 20 000 to 25 000 foot hand for NS, L. - 1. 5, and the most sensitive altitude band decreases with increasingKs. L to the 10 000 to 1.5 000 foot band for KS. L. - 6, 0. The calculation of the most sensitive band is treeated mathematically in Ref. 5-I. Tne ballistic coefficient p varies with altitude and Mach number. Tne value of P at hypersonic speeds should be usrd for this calculation (also for 8 discussed in the next subsection), The summation term in Eq. (5-1) is called the ballistic wind. By definition, a ballistic wind is a constant wind (with altitude) such that the dispersion is the same as that resulting from a variable wind. If Vw is constant at all altitudes, then Eq, (5-1) becomes:
V H

,,

8M,

VE siin

1:3

(5-2)

The quantity p)..eceding the summation sign in Eq. (5-1) is called the range partial for Ainds. In Ref. 5-1, It is shown that Eq. (5-1) is valid for either head (or tail) winds or cross winds. If the wind

-283-

r..
S.I

TNS JOHNIS MIMPAINS LJNiVORSrfY

APPLIED PH4YSICS LABORATORY


6MVIR SPRING UAR,..AND

is neither a head (or tail) wind nor a cross wind, the total dispersion may be obtained by computing the head wind and cross wind componerits of 6wand ad ding the results vectorially. Equation (5 -1 ) has been found to be in excellent agreemetht with results obtanined using three- or six degree -of -freedom trajectory simulations provided KS. L. is not gre-tter than about 6, For higher values of K8 L the equation overestimates 8 w. For applications where the body characteristics have been fully defined, the accuracy of Eq. (5-1) may be improved by the following procedure: 1. Using a three -degree -of -freedom trajectory prograin, evaluate 8 w for, various values of VE and ^/E and a constant wind, Using these results and Eq. (5-1), 13 and plot it as a function of r. compute

2. 3.

Using the trajectory program, evaluate aw resulting from a constant wind in each altitude band. These computations are Made for one value of yE and the lowest IVtEI to be used for the missile. Using the results of step 3 and Eq. (5-1), compute Fw*

4,

These new values of 13 and Fw will differ slightly from those given in Tables 5-1 and 5-2. Using these new datai, the accuracy of Ow given by Eq. (5-1) may be improved,. . particularly for high values of Ks 5. 1. 2 Effect of a Deviation in Density The impact point of a reentry body is computed using some standard de-nsity profile (variation with altitude). If the actual profile is different from the standard

-284-

41

APPLIE'D PHYSICS LABORATORY


SILIO1 SOlNG MANVfLAND

profile, the missile velocity history is affected. of impact dispersion.

Therefore,

deviations in density from the standard profile are a source.

Using an analysis similar Lo that used for 8w, is shown (Ref. 5-1) that: 8,0 : 85}

it

M VE sinE

Y tan7

1 E N 5

&P5 F 0j

(5-3)

where 15 is given in Table 5-3 F is given in Table 5-4 44 is the deviation in density from the standard value (p) at a given altitude. All other terms are as defined in the previous subsection on the effect of wind. Again, F is a weighting factor that depends upon The most the altitude, altitude bandwidth, and KS. L. sensitive altitude band is the 25 000 to 30 000 foot band (see Ref. 5-1 for fuller discussion) and is nearly independent of KS, L.'

The summation term is called the ballistic density. By definition ballistic density is the constant value of ap/othat results in the same dispersion as the variable The quantity preceding th'a summation sign is called the range partial for density deviation. Equation (5-3) has been found to be in excellent agreement with results obtained using three- and six-degree-of-freedom simulations, provided KS L is not too high. In Ref. 5-1 it was shown that the condiftons rk.quired for the accuracy of Eq. (5-3) are those given in Figs. 5-1 and 5-2. When these conditions are exceeded, the dispersion given by Eq. (5-3) is too high.
IL

-285

tS{_

t149 JOHN% "OPRtNM UPt1IVtkfgh

APPLIED PHYSICS LABORkTORY


SIJIM SlPIINd MAIRYLANL

For applications where the body characteristics have been fully defined, the accuracy of Eq. (5-3) may be improved following the same procedures given for winds in the preceding subsection. Since values of F. and Fw become very low at high altitude, one is tempted to conclude that the dispersion resulting from winds and deviations in density at altitudes above, say, 100 000 feet is negligible. For a high precision vehicle this conclusion Is not necessarily valid, since the magnitudes of Vw and A"/p at these altitudes are usually very large. It should be noted that, except for the direct effects (Ao/o and Vw), the variables contributing to dispersion resulting from winds and deviations in density are the same, However 80 is more sensitive to changes in the variables. 5. 2 Errors in Aerod ynamic Coefficients Errors or variation in aerodynamic coefficients res;ulting from reentry environmental effects (such as heating) may contribute to dispersion. The most sensitive coefficient from the dispersion standpoint is the axial force coefficient, CA.
I/ history during reentry,

Dispersion results when a perturbation affects the


This change in ' was shown

(Ref. 5-1) to result primarily from changes in the K hibiory.

In Section 5.1.2 we interpreted changes in K in terms of deviations in density. We could just as well have interpreted a variation in K as a variation in f3

Sicl

thenand

PAS

the__

II

and the Wc thenAK A Ws E dispersion characteristics computed in Section 5.1.2 are The equally valid for ,CA/CA, - AW/W, or - Ad/1, minus sign means that a given percent increase in W or Sresults in ae same dispersion as the same percentage decrease in pressure, density, or axial force coefficient. or sinYE. Since

-286-

I>,

I
THK JOHN$ HOPKINS UNIVSNSSITY

APPLIED PHYSICS LABORATORY


I.VSR @PRINS MAIIVL.ANb

Errors in the normal force coefficient have a small


effect on 8. Since in general the angle of attack is not zero, a lateral. force results, and the magnitude of this force for a given trajectory depends upon a and CN,. The dispersion resulting from this source arises from erroneous evaluation of lift and drag forces (see Sections 5.4.1, 5.4.2, and 5.5). This increment in dispersion is very small unless the error in CN or the angle of attack is large.

Errors in the damping derivative may result in


dispersion since this characteristic determines to some degree the 0. history and thus the lateral force history early in the reentry phase and during periods of roll resonance (see Sections 4, 4 and 4. 5. 5), Again the dispersion resulting from these errors is usually small. Errors in the center of pressure location may result in dispersion since this factor also determines to some degree the ;. history early in the trajectory and the a resulting from asymmetries. The dispersion contribution is usually small. 5. 3 Errors in Mass Characteristics Errors in mass characteristics that contribute to dispersion are errors in W, Xc g. , and the moment of inertia terms. Of these, the most significant is the error in W, which results from uncertainty in the reentry weight and uncertainty in weight loss caused by heating effects during reentry. The dispersion resulting from an error AW/W is the same as that from an error, -"p/p, as discussed in Section 5. 2. and moment of inertia may The effects of rc. cause small dispersions sl'nce these factors affect the 0-i history during reentry. A

AK

287

.K

.,.

_ __ __ n m ~ n _p

m m

p - i u uv ~ n r i n

:1. , au

i n

THR JOHN$ WIOOIND UNIVI5ITr

APPLIKID PHYSICS LABORATORY


5VLYR SNPI4N0 MARLAND

5.4 Asymmetries
The problems associated with asymmetries were discussed in detail In Section 4.5. Asymmetries produce angles of attack; angles of attack result in lift and n change, in drag. For cases where ingles of attack develop and the missile is rolling, the major effect on 6 is the change in drag since the direction of the lift vector (in an inertial axis system) is constantly changing so that, over a period of time, the net effect of lift is "averaged out. " The effect of * for a rolling vehicle is usuallyjsmall unless resonance conditions occur. In this case o may become large and the vehicle drag will increase significantly, causing the missile to fall short of the intended target, This case can be treated as a change in CA so that the

procedures given in Section 5, 1, 2 may be used.

A more

serious contributor to dispersion is the case where 0 and the roll rate becomes zero, Even for a very short period of zero roll rate, large dispersions may result from small angles of attack, Since, for this case, the lift vector remains fixed in space, the vehicle moves off course in the direction of the lift. We will consider these two cases separately. 5.4.1 Rolling Vehicle with Angle of Attack

If asymmetries result in a trim angle of attack and the missile roll rate is not zero, the major contributor to dispersion is the effect of the increased drag resulting from . The equation for the drag is: D =C qS (5-4)

wh-ere, from Eq. (2-9): CD CA cos &,+ CN sin . (5-5)

Therefore the change in CID resulting from O is,:


&CD

CA cos'+

CN sin

- CA

CN sine

(5-6)

-288-

A-'

APPLIED PHYSCIS LAIORATORY


*OLVI *llPil. MAT&Pi.N

THE

JOHNi

HOIPKINS UNIVtUITY

If the aerodynamic characteristics are linear so that


SCN - CN ai and if a is small:

4 CD = C N

(5-7)

Therefore the effect of i may be estLmated using the procedures given in Section 5. 1, 2 where:

ICN
A A

(5-8)

W'When working with telemetry data, it is convenient to express ACD/CA in terms of body accelerations, Substi-

tutIng for & using Eq, (7-11): A C Dc A 2

(A()

Note that ACD/CA varies as the square of

&,

and there-

fore the effect tends to be small at low values of & but


tends to increase rapidly with increasing e. 5.4.2 Nonrolling Vehicle with Angle of Attack

"J

-close

The dispersion resulting from tngle of attack for a nonrolling vehicle was considered in detail in Ref, 5-2 and the results are summarized below. Consider a vehicle with a fixed angle of attack that results in a lift force L, Consider an inertial axis system YZ as shown in the sketch. "Ifthe roll rate is low compared to the aerodynamic frequency, the lift (or normal) force is in the plane containing the asymmetric force (see Section 4. 5. 1). Therefore "L rolls with the body. As long as the roll rate is not to zero, the effect of L on dispersion is small since the direction of L (in the inertial axis system) Is constantly changing. However if the roll rate becomes

IL

- 289-

il

THK JONH46 HOPKIN

uNglVKOIr-

APPLIED PHYIiICS LAIDORATORY


11ILVI %PoliN$ MARYLAND

Im I

ZY

.1
zero, even for an instant, the body will move off course in the direction of L. We wll consider in this subsection the case of a vehicle that is spinning at reentry but, under the influence of body asymmetries, the directior of roll rate is reversed during entry, The duration of "near zero" roll rate is very short. Assume that at b the roll rate passes through zero and that at this instant the lift is in the vertical plane. A change in flight path angle, A', results,

I-ORIZ.ONTAL H..

P'
19I NEGATIVE

a'?
tN

ISPOSITIVE

29

-::

....

.,.

-... , . ..

'

Thi JONNiI

MOP4iN3

uNIVll6ItY

APPLIKID PIHYSICS LABORATORY


MILYIS3PRIENG MU VLAkND

Assuming a straight line flight path and a small value for

.2......(5-1

'

in2 s in

0)

L Is in a plane SIf perpendicular to the vertical plane and if AV is interpreted as the resultant change in the direction of the velocity vector: h AV S.18

sinYl VE

(5-11)

SI

"

The impact pattern is elliptical with the major axis along the trajectory track and the minor axis across track. The ratio of the minor to major axis is I sin YEl'. The change in Y is given by the equation: Ay .dt (5-12)

where (see Reaf. 5-3):

g WV

COSY + V cosy V

The change in V of interest is juqt that resulting from L. Therefore: AV 9

Ldt

(5-13)

Equation (5-13) is valid provided the direction of L is fixed. However, since the direction of L changes over the time interval of interest, an effective value of .r Ldt must be used. Based on six-degree-of-freedom simulation results, it may be shown that:

- 291-

-li

APPLIED PHYSICS LABORATORY


MILVt* SPKlNG MAU IAV.AN-

!I4R JOHNI HoPKINS UNIVEftSI1Y

2.11 6L.
Ldt 2

IrT I
Writing L

(5-14)

where the constant is dimensionless, where C is given by Eq. (2-10):

CL OoqoS

fLdt

(CN

- CA)

qS

(5-15)

Substituting Eqs. (5-15) and (5-13) into (5-10): 2,116hg (CN R = WV sin 2 E CA) I 0 q0S. (5-16)

In terms of normal force No 2, 116 hg


= 0

CN

q S: (5-17
N (5-17)

WV

sin2
-2
0 Io

In terms of lateral acceleration g

+ gZ

2
0

N0

. 116hg 1 0(

( CA\
CN-8

V0 sin ^141pof: Equation (5-18) is convenient for analysis of telemetry data since lo and go are obtained directly from the data and Vo and h 0 may be estimated using Eqs. (4. 3-22) and

0I

-292-

.0,.

I
YP[ J1M3HNIMOPKINS UNIVIURMSIT

APPLIED PH4YSICS LABORATORY


l,,.VllIP *

M.RAm'tNB l@NG

(4. 3-5).

Equations (5-16), (5-17), and (5-16) are valid for

any combination of asymmetries, We consider next the dispersion that results from the particular types of mass and aerodynamic asymmetries (N and r) shown In the sketch.
0r

1:1

AXES) V (BODY

No

We will assume that the normal force resulting from r is very small so that the No results from an asymmetric moment defined by: no -- m 0=C2+C2s "

The trim angle of attack, ao' is given b_ Eq. (2-27) where, in the notation of the present section, CITR a ao and Cm. Z!M .
0

The roll acceleration, O, is given by Eq. (4, 5-:35)

qo, atocio, and 0 - 0o. Substi'uting these values where q into Eq. (5-16), writing po In teorms of H and Pop and noting that q 0 POIVO

-293-

S.

:.. ,., I ,.. :,: .: ,. .. .. : ... .. ::) .. ... .... S.

.....

.... .......'.

"TiL JOINII

NOPKINS

JNii.MSITV

APPL.IED

PHYSlC5 LADOOATC)PY

1.4 r

- C N /g w ,i ' -

SI X sP

1/2
0

(19)

The constant is 2, l11/fTand is dimensionless,

Equation (4. 5-94) may be used to evaluate Ko and thus ho 0 o. Using Eq, (4. 5-94) and Eq, (2-27), 6 1 may be expressed in terms of io rather than Cm and r sin 0o, the result:

Sgiving
2'

A
6

1. 49i
1. 49oE 0

Po

sill Y

S." L. (5-20)

Equation (5-20) shows that for a given o (given Cm and 'AX/d primarily), the dispersion varies with h. (vartiations in r sin 00 if other variables are held constant) in accordance with the factor: oP K
0

S.
2 =f(h K (5-21)

P E h S. L. I e

PS. L.

At sea level, this factor is zero since h,) 0. The factor increases with increasing h. to a maximum and then decreases, becoming zerc at very high altitudes. The criti-

cal altitude (altitude for:

-,uximum E~) depends upon the

type of atmospherc WS. L.) and upon KS.L. (or upon and/orYE). For the exponential atmosphere (Eq. (3-15))

294

APPLIED PHYSICS LABORATORY


SILVER 1 PRINI 4 MW43TI.AN

it may be shown that E is a maximunr at ho Ho for very small values of KS L, and increases to ho = 2Ho for very For the 1962 Standard Atmosph-re, large values of KS using 11 = constant' -3 000 feet, the critical altitude varies from about 23 000 feet at low KS.L. to about 46 000 feet at A plot of E versus ho is shown for several. high KS. L.. values of KS L. in Fig. 5-3. In summary, the following conclusiors were reached regarding the dispersion resulting from reversal of roll rate. 1. The dispersion may be in any direction. The of possible impact points is an ellipse centered at the locus nominal impact point. The major axis of the ellipse is along the flight path. The ratio of minor to major axis is

sin y.
2. The magnitude of the dispersion is a function of the many factors Listed below, making it impossible to draw generalized conclusions regarding the effect of specific variables. CA CAC CN m As Asymmetry

Noo

Aerodynamic

c.p.

W
X (Mass PE Initial Conditions

)
! Geometry

Type Atmosphere

S
S

I
-

295

-i .i

- ,

,.

"

V.II i

.... f."

t-.....

]l

T1I" JOMPl2 NOPKIN5 *ILfU *p*,NO

UNIVKSIRITY

APPLIED PHYSICS LASORATC)O


M*VANtI,

Only one type of combined asymmetries capable of resulting Ln & ind zero roll rate ir listed nabove. Hlowever, other combtined asymmetries such as C, and Cm o could also result in roll L-ate reversnl and Dirge dispersion. 3. For given nexmrovmyn:inic kI ielutcistics, mass, g.eometry, initial conditions, type of atmosphere, and Cmo, the Wo is fixed, However, h. (and No to a very small extent) varies with the values of r and 0o" The dispersion that results as r and/or 0 is changed is maximum when ho has some specific value. For the 19f12 Standard Atmosphere, the value of ho corresponding to 6 max is between 23 000 and 46 000 feet, depending upon the value of Ks. L. 5. 5 Errors in initial Reentry Angle of Attack and Body Rates

In some cases the initial reentry angle of attack, "i, and body rates, p, q, and r, are not known accurately, During the high altitude portion of the trajectory, these initial conditions affect the e history as discussed in Section 4. 4. If W, p, q, and r are not zero in the altitude r'-gion where the aerodynamic restoring force is small., the missile experiences coning motion such that, in general, the body centerline does not rotate about the velocity vector. Therefore, in inertial axes, the lift force does not rotate about the velocity vector an it does for a rolling vehicle at altitudes where aerodynamic forces are predominant. During the early phase of reentry, initial values ofrs, p, q, and r cause dispersion resulting frorri both lift and drag effect,. The drag effect always tenois to cause the vehicle to fall short of the intended impact point. >lowever, the lift effect may cause dispersion in any direction depending upon the direction of The lift vector (determined by the location of the coning motion relative to the velocity vector), In general the locus of impact points resulting froi i those initial conditions is an ellipse whose major axis is approximately along the flight path. The ratio of the minor to major axis is approximately Isin Epl. The drag effect shifts the center of the ellipse toward the launch site from the nominal point (impact for nominal y, p,)q, and r).

I296

S...

HiK JOMNP49 NPRINN UNplvit"ItiI

APPLIED PHYSICS LABORATORY

3I~lIVN
SP*IN6

MARIL&ND

!
The vehicle motion is rather complex, and a six-degreeof-freedom simulation is required to evaluate the magnitude of this source of dispersior', A typical order of magnitude of the semimajor axis for 6 resulting from ;; is 5 to 10 feet per degree error in 9. Therefore, the error is not negligiblc unless W is known fairly accurately. 5.6 Examples 5. G. 1 Example 1 Estimate as a function of E the dispersion resulting from a constant 100 ft/sec headwind (V -100) and a constant 10% increase in densLty (Ap/p 1.) from the 1962 Standard Density for the following conditions: V E2 20 000 ft/sec 1= 000, 200n, and 3000 lb/ft Since dispersion is most sensitive to the altitude region from 20 000 to 30 000 feet, a value of H 23 000 feet (see Table 3-1) is used. From Eq. (5-2): (-100) x 000 w 23 000 sinE 3 w20 ,,I

15 sin3 y

From Eq. (5-3): 8=~ 6 123 000 20000 2 32.2x 0.1 x 15 2 sin V.Etan VE 4. 26 5 sin VEtan7E L. and_8
.
:1

From Table 3-1, P 2116 lb/ft 2 . Then K, 2116 6 For a given value of 0 and VE, 8, #sinsi~ yE

may be computed using 13 and 15 obtained from Tables 5-1 and 5-3. The results are shown in Fig. 5-4. This
- 2072

-297

b-----__

THK JOHNSHOP1,INI UNIVKRSITY AhPOLED PHYSICS LA3I0RATC)RY 1BYIR 801110G MARYLAND0

figure illustrates the high sensitivity of dispersion to winds


low values of /. and density for shallow reentry angles and at It also Illustrates the fact that 8 becomes insignificant for even large values of )E whereas 6w may be significnnt

5.6,2

,xample 2

Assume that in targeting a mnissile no corrections than 1962 StandardDensity profile for altitudes greater from 100 000 100 000 feet. If the average wind at altitudes if the denand headwind ft/sec to 160 000 feet were a 300 profile, Standard 1962 the sity were 30% greater than of 3 the impact error resulting estimate as a function from these high altitude meteorological characteristics for VE x 20 000 ft/sec and YE From Eq. (5-1): -25n. Use H1 23 000 feet.

20 '000(-0, 422) 4 -B17 I

23 000(-300)

4
.3 rj2WF

'I
44

F.

From Eq. (5-3): 23 000 0.32,2x

Ij~
iI

6~20 000X-0
4
=-155 15

422),

-7765

34

r'

,! 2

E, Fw,

Using values of 13, Fw, 15, and F given Inr were computed ns a Tables 5-1 through 5-4, 6w and 8

w 0
-298-

THN JOWNN "OPKINS UNIVdSKSITY

APPLIED PHYSICS LABORATORY


SILVI

SPO.NG

MARYLAND

function of KS L.T The values of corresponding to KS L. were computed using the equation: 2116 5000

KS. L.

KS. L.

Values of 6w and 6 are plotted in Fig. 5-5. The accuracy limit as given in Fg. 5-1 is also shown. Values of z: 1230 lb/ft 2 fall within the limits of accuracy; va.ues of 8 shown for p less than 1230 lb/ft 2 will be slightly high. This problem illustrates that bodies having low values of A tend to be sensitive to meteorological characteristics even at very high altitudes. 5. 6. 3 Example 3 For the reentry body havhLig the characteristics given in Table 2-1, estimate the dispersion resulting from a constant trim angle of attack ,if the missile has a con-

istics are VE

stant roll rate during reentry. The trajectory characterUse H = 20 000 ft/sec and YE = -3023 000 feet.

For this case the dispersion results from the increase in drag, and the effect on dispersion is equivalent to a change in density. CN

(Eq. (5-8))

C 0' *o2
=A 2
0

0. 104 *

-2 o
0

.- 2 fore =19.2
0 = 0

in radians
0

0. 00585 E 2 for W - 299


-

in degrees.

,.i

?H9 JOMNO HOPKINB UNIVygRMIT1

APPLIED PHYSICS LABORATORY


SILVIR SP09046 WAnVANP

The dispersion is computed using Eq, (5-3). - 2116 1000 (-0.50)

S. L,

15 = 21 Since A0I N is constant for all altitudes:

& F

-0. 00585

-2

Therefore:
=

p 20 000 (40, 50)


-36. 2
-2.
0

200 23 000 2

)2

32,2 x 21 (0.00585)N -2 2 3.2x2 0 (-0. 577)

The missile will fall short bL9 feet for o= 1/20, 36 feet for &o = 10, and 145 feet for &o = 20. 5.6.4 E_.xample 4 Suppose in Example 3 the reentry body has an jo = 1 / 20 at reentry, the body passes through first resonance without lock-in, but the asymmetries arj such that lock-in occurs at an altitude of 10 000 feet and remains until impact. Determine the dispersion. Again, for this problem, dispersion results from the drag effect. However the angle of attack is amplified
by the resonance phenomena, From Eq. (4. 5-16), ,

S= Ai

and, since X - 1 at resonance, A Eq. (4. 5-60):

1/D.

Using

0. 115

lr'W.

-300

'I

THE HOPKINS UVdIVINSITY Tie JIOHN$ O~lHPIIt~VRIAPPLIED PHYSICS LABORATORY

Therefore: 8.7(1/2)
-

4._35

Substituting for K

P sin

P E PS. L.

KK

4.23

. .2,11

L..
At h = 2500 feet h = 7500 feet P/ P/P 0.0913 0.757 2. 21

S. L.

a = 2.420

From Exalple 3, the equivalent value of A&/0 is 0. 00585&o . Therefore, for h = 2500 feet, AL/p = 0. 0286; for h = 7500 feet, Ano/p = 0. 0343. Assuming that the value of 60/p at h - 2500 feet is representative of the altitude band from sea level to 5000 feet and that the value of 6p/p at h = 7500 feet is representative of the 5000 to 10 000 foot band:

17
F j z-1 6 Using Eq. 0 (5-3): 3-2.577) x 21 x0.00329
= 0. 0286(0. 049) + 0. 0343(0, 055) =

0. 00329.

20 000x0.5

-20 feet

5.0.5

Example 5

The body having characteristics given in Table 2-1 ias the following asymmetries: 1. A center of gravity offset of 0. 042 inch.

S301
I I I I I "

Tir JOHNS HOPKINS UNIVEIRIITY

APPLIED PHYSICS LABORATORY


SILVER SPRPNa MARVLAND

2.

A body centerline distorted into an arc of a circle. The slope of the centerline at the body nose and base differ by the angle A&.

As a function of A&, assess the probability of encountering spin through zero roll rate, and estimate the dispersion that results if spin through zero roll rnte occurs. Neglect the possibility of encountering roll lock-in at negative roll rate. Use the 1962 Standard Atmosphere with initial con-30. ditions, PE I 1000*/sec, VE = 20 000 ft/sec and From Eq. (2-35), Cm (2-27): 57. 3 (0. 008A 8) =2&8 , 0. 0108 A8. From Eq.

0o where AG and

2(0,114)

, The altitude (in terms

O are

in degrees.

of K) at which the roll rate becomes zero is given by Eq. (4. 5-94). Substituting for the given terms: 1000 -; ~K/2 B (0.104)(0, 079) 20.001)( 3. 5 0.36 -i
- 3 sin 0 57.

9 sin 0~
4. 23, For

2116/1000 (0. 5) At sea level K : K zero roll rate at sea level: A sin 0
01-

0.36 - 2,115

Since 0o may have any value from 00 to 3600 and assuming no correlation between the two types of asymmetries, the probability of encountering zero roll rate during reentry is:

3
U

-302 -

I
THR JOHN$ HOPKINS 61NIVCMSItY

APPLIED PHYSICG L.ASORATORY


616VIR SPOING MARYLAND

I0

AO

where A0o is the range in values of 00 for a given AO such that

8 sin

. 41
0

/,esin0
0.41 0. 50 0. 60 0. 70 0. so 1.0 2, 0
S. 0

0I Ao 0.O
0

1 0. B2 to 1 0. 683 to 1 0. 585 to 1 0. 512 to 1 0.41 to 1 0, 205 to 1


0. 137 to 1

0 70 94 108 118 132 156


1 64

0 0. 195 0. 261 0. 300 0.328 0.367 0.434


0. 456

The probability of encountering zero roll rate is shown as a function of Ae in Fig. 5-6. The altitude, ho, at which zero roll rate occurs is an important factor in the calculation of dispersion. For- Ae ; 0. 41, ho may

be any value from sea level to some upper limit obtained when sin 00=1I or

The upper limit to ho is shown as a function of A


5

in Fig.

The dispersion that results when spin through zero

roll rate occurs may be computed using Eq. (5-20). For a given Ae (or %O), the dispersion may be any value from 0 to some maximum value depending upon ho which, in

i..

"
ii

-303-

- 3oW.

Twit Jo0N$ 0OPRING UNIVINNITY

APPLIED PHYSICS LABORATORY


SILVENq SpaNo~ UAA~I.ANDb

turn, depends upon 0o.

For low values of A6, the maximum vRlue of E (Eq. (5-21) and Fig. 5-3) correspond to the highest value of ho (00 = 900). However for the larger values of A6, the maximum value of E for this problem (KS. L. - 4.23) occurs at hof 27 000 feet. Using ho data given in Fig. 5-7 and the E values given ;n Fig. 5-3, Emax for various values, of A8 are taoulatud below, h 0,e atE maX
0.41 0.5 0.6 0. 7 0.7 1.0 2.0 3.0 0 13 000 21 000 27 000 27 000 27 000 27 000 27 000 = 2 Ae: E max 0 8 600 10 800 11 300 11 300 11 300 11 300 11 300

AeE max
0 4 300 6 480 7 920 9 050 11 300 22 600 33 900

Using Eq. (5-20) with&,

" 4(2AO)

(0.104 20 000 x 2116 .max 11000 x 23 000 x 1000 (1/2)2

max

m R maxe

1. 22A 9 E max .

j
"

The values of 8 Rmax may be computed using the listed values of Emax, and the results are shown in Fig. 5-8.

The impact may be anywhere on an ellipse having


a semimajor axis of any value from zero up to the values shown in Fig. 5-8. The major axis is along the trajectory, The semiminor axis is 8R/2 (see Eq. (5-11)I.

-304-

_.

________________,... ...

::::'?

-. ,,,;,:.:

, .

I
I ?M| JOgNSm HOPKINS UNIVIRthIY APPLIKD PHYSICS LABORATORY
IlLV911 ISNING MAPYLAND

The magnitude of the dispersiun is highly dependent upon 00. In the previous discussion we considered only the maximum effect. The probability of encountering a given magnitude for the major axis of the dispersion ellipse may be obtained by plotting 61 versus 00 for various values of A as shown in Fig. 5-9. The dispersions for values of 0o > 900 are not shown since OR is symmetrical about 00 i 900. The probability of encountering or exceeding a given value of ORfor a given A6 is obtained from the ratio A0o/360 where A0o is -the range in values of 00 for which 8 R equals or exceeds the specified value. The probability plot is shown in Fig, 5-10, The values of P shown for 5R = 0 are the probabilities of encounterIng zero roll rate shown in Fig. 5-6. As an example of the interpretation of this figure, for A8 - 0. 50 there is a 19% probability that zero roll rate will be encountered; there is a 16% probability that 8 >- 2500 feet; the maximum possible dispersion is 530Ofeet (the values of OR at P' =0 are the values shown in Fig. 5-8). Note that the dispersion that may be encountered if the missile spins through zero roll rate may be at least an order of magnitude greater than the dispersions considered in the previous problems, For example, the A& = 0. 50 asymmetry corresponds to a 10 trim angle of attack. It was shown in Example 3 that if the missilc does not encounter zero roll rate the resulting dispersion is about 36 feet for a. = 10. The spin through zero roll rate phenomenon is a serious problem to the designer of reentry vehicles that require good impact accuracy.

' .

...

-305.

"

I
?rHe
JOHNS HOPKINI UNIVUlNsIY

APPLIED PHYEICS LABORATORY


B1IgvRRSPRINd. UAMILAND

REFERENCES

5-1

L. S. Glover, Approximate Equations for Evaluating the Impact Dispersicn Resulting from Reentry Winds and Deviations in Density, APL/JHU TG 1132, September 1970.

5-2

B. F. Fuess, "Impact Dispersion Due to Mass


and Aerodynamic Asymmetries,' J, Spacecraft and Rockets, Vol, 4, No, 10, October 1967, pp. 1402-1403.

5-3

W. H, T, Loh, Dnamics and Thermodynamics of Planetary Entry, Prentice-Hall Space Technology Series, 1963.

I.I
..:

I.
307
-

I[ :1 .;..

Preceding pag blank

. . . ........ .,

,: ..!

THI JOHNI HOPKINS UNW9*11ITY

APPLIED PHYSICS LABORATORY


SILVI SPRIN6 MARYLAND

!7

20 18 10 14

12

10 VELOCITY (thouem.do of ft/,'eol

18L " 2116lb/ft III 100000feet O H - 23000 teet

TAN A 0.2 TANy E

0
0 -10

,
-20

I
-30

i
-40
7'

I
-50 (degrees) .-60

I
-70 -80 -90

Fig. 5.1

LIMITING VALUES OF K .L

Preceding page blank

-309-

TH

r ,IOHINI HOPKINS UNIV

lSITYV

APPLIED PHYSICS LABORATORY


SILVIO
SPRING MAPLANP-

-60

T r SPS.L,=
hi H

2116 Ib/tO

i-

PS

100 000 feet Ln "P - = 4.5) 23 000 feet A-= 50

STAN
-40

02500

-30

1000
-20 2000 --3000 -10
00

10

12

14

16

18

20

22

24"!

V E (thousands of ft/sec) Fig. 6-2 LIMITING VALUES OFyE

-310

'

"

!NK

JOHN$

k.

)PKING UNVlRSIyTy

APPLIED PHYSICS LABORATORY


5Im.vim spimwe MAIYLAND

20
oat

I hh
K *MaxA -O

1 AT I ATE rKn ATKsL-. EMPERATURE CONSTANT

TEMPERATURE

K12SL,

4'

6
4
2

10

20

30

40

50

s0

70

80

ho (thouuandt of foot)

Fig. 5-3 VARIATION OF E WITH h. 11962 STANDARD ATMOSPHERE

-311-

V049 JOHNS 1ilVI.

'OPKINU UlNlVtU51V SPINO MAIPLAD

APPLIED PHYSICS LABORATORY

-10 000

-1000

-1

sw (100 ft/imc
z 6p (10% INCREASE)

-oo
-100

"
I

2
__l

000.
I

30000

-20

-30

-40

-50

-60
(dogrees)

-70

-80

-.90

"7E
Fig. 5-4

DISPERSION RESULTING FROM WIND AND DEVIATION IN DENSITY

-312L/

4H

THE JOMHN HOPKINi UNIVERSITY

APPLIED PHYSICS LABORATORY


9OL*"m SPRING MAQVYLAOI

-500I

400

(300 ftWas

HEADWIND)

8p 130%DENSITY INCREASE)

5+5)

J'30
.200

\\SA
\ ,
LIMIT FIG. 5.1 L ACCURACY

Iw

A
1800 2000 2400 2800 3200

80o

1200

~3(Ib/ft) I
Fig. 5.5 DISPERSION RESULTING FROM METEOROLOGICAL CHARACTERISTICS AT HIGH ALTITUDE

-313-

4k,

APPLIED PHYSICS LABORATORY~


GLVRM BPF41MGMARVhAND

NN

I
jN

0 03

pa

314

IMIL-'*

APR,.IED PHYSICS LABORATORY


OpeN MANYLANO Gb

tHS JOHMN 14@PKIMNE IIYEMITY

0-

~IN

US

jiselospuno~i 04wwixv
-IF
c

I-

315

APPLIED PHYSICS LABORATORY


416VIll

TIHg JOHNS HOPRIMNSUfNIVWRSITV

MANVLANIP UPOING

u:1

311

SS&.VUl

APPLIED 16YICvISa LABORATORY SO1M, MANVLAND

IC

000

dVd

7-

1.1

Z. ?H JONI" 11 APPLIED OP(INSLABORATORY UNIV0II2tY PHYSICS


Il.LyV
5PIJINMS

MAM.AND

tt

0,15

0,: - 3 .0 AO

0.3
a.

ol

0.2

0.50
V I.?

0.1

10

20

30 (thausarnds of foot)

40

50

Fig. 5-10

DISPERSION PROBABILITY

I.I.,

318-

. . . . . . . . . .. . . .... . . ...,. . . ,.. . . . . . .. . .'. . .. . . . ".

T49

JOHNS

14OPA0INO

UNIVIERIY

APPLIF.D PHYSICS LABORATORY

Table 5-1 Values of 13

KS. L. 0 0.5 1.0 i.5 2. 0 2. 5 3, 0 3.5 4.0


4,

13

0 0.25 0.57 0.92 1. 33 1.84 2,44 3.20 4.13 5.25 6.60 9.90 :.

5.0

5.5 a8.17
6.0

319-

Ii

II.

APPLIED PHYSICS LABORATORY


016VBR S*PRNN AMARYL&ND

-Altitude,1L
(thouands l . o ....

t.....
1 . 5 0, 007 n 2 3 4 5 (i 7 H 9 10 11 12 13 14 15 1, 17 0. 005 0.010 0.025 0,021 0, 030 0, 045 0, 0134

Valjuc,

nf IF

2. 0

3. 0. 005 0, 004 0,007 0,018 0, 011 0, 021 0, 030 0, 047 0,.

0 0, 004 0, 002 0, 00 0,013 0. Ol1 0,017 0, 024 0. 033 0,053 0. 077 0,110 0. 071 0, O((i 0.105 0. 1.12 0. :10

, 0, 001 0, 002 0, 005f ol011 0. 0091 0. 015 0. 020 0. 029 0. 042 0. 066 0.100 0. 068 0. 085 0.114 0, 15B 0. 1.16 0.127

, 0 0, 00:3 0,001 0.00.I 0,010 0,0(t 0.013 0, 0161 0. 025 0. 0:4 6 0. 0113 0,0!17 0. 0U 13 0,0 1(G 0.117 0, 1[ill 0.1) 5 0. 1:12

0. 00Cl 0. 004 0. 009 0.023 0, 019 0, "27 0, 040 0, 058 , 071 0,.1 O 0,143 0. 08(1 0. 100 0.0. 90 0. 080 0,070 0,0 900

140-I (10

S20-140
100-120 90-100 80-90 70-80 ((-70 -60 40-50 30-40 25,-3 0 20-,5 15-20 10-15 5-10 S. 1. -5 Ntue.

O
0, 101 0(,147 0. 081O 0, 103 0. OB6 0. 0119 0, 0,8 0,047

ou0n

0, 0'2 0. 1261 0. 078 0. 0112 0. 097 0. 110 O.101 . O92

0. 11.1

T1 E'sL. data are applicable to a 1962 Standard Atmo)spher'e. Ilowuvr, th..v Wi v be used for auin, atmosphere by chnngillg the altltudc limits for ench h-and. 'r' example, the .1 dantn npply to the altitude band from nn altitude corrusponid1mv to PP L . 0714 (610 000 feet for the 19102 Standard Atmompherc)to an altitude coript'e ncsPL ttO PIPs, L. 0.0443 (70 000 feet for the 1602 Stnndard Atmosphere).

320-

_,I

TMS

JOHN*

UNIVURII0|TY 8OPKINS

IABORATORY PHYSICS APPLIED 3t ,..lO4 MARIIhANb IIIIvg6

Table 5-3 Values of 15

KSL

15

0 0.5
:

0 0.50 1.05
i. 85

1.0 t, .5 2. 0
2.5

3. 00
4.90

3.0 3.5
4.0

7. 55 11. 30
17 30

4.5
5, 0 5.5 6.0

27. 00
41.00 61.00 92.50

S321

"ti"

THE JOHI"N$H'OPKINS Uhl[CllllfV

APPLIED PHYSICS LABCRATORY

Table 5-4 Vanu#, of F A lttude (feet x 10 O ) 160-140-160 120-140 100-120 90-100 80-90 70-80 60-70 50-60 40-50 30-40 25-30 20-25 15-20 10-15 5-10 0-5 p
KS . L.

_....

j
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

1,5 0. 004 0,014 0.027 0.053 0.045 0. 055 0. 069 0. 088 0.118 0.142 0,140 0. 072 0. 068 0. 044 0.030 0.019 0.012

2.0 0. 012 0.014 0.023 0,046 0, 039 0. 050 0, 067 0, 085 0,111 0,138 0.146 0,076 0. 074 0, 048 0.032 0.022 0.017

:3,0 0. 009 . 00;) 0.017 0.035 0.028 0, 039 0. 058 0. 077 0.099 0.130 0.154 0. 083 0.082 0. 064 0.051 0.037 0,028

4.0 0,006 0.006 0.012 0.025 0.021 0. 029 0.045 0.065 0.090 0.123 0,157 0. 088 0. 090 0.078 0,066 0.053 0.046

5,0 0. 004 0.005 ,. 008 0.019 0,015 0. 025 0, 034 0. 056 0,087 0,120 0,158 0. 092 0. 098 0. 087 0.076 0.061 0,055

6.0 0. 004 0.004 0.007 0.017 0.013 0. 022 0. 030 0. 051 0. 086 0.118 0. 156 0. 095 0, 102 0. 091 0, 080 0.065 0,059

Note: See note to Table 5-2.

-322-

I
tHE JOHNS NOPI"IH INYiMRMITY

APPL19D PHYSICS UORATORY

I I I I !
1
6. TRAJECTORY SIMULATION

I II

I SI

I-

323-1

I
i aa

YNS J@HNsMOM",N UWIVSUI?

APPLIED PYSICS LABORATORY


hILYSN uP*IN* MASYLAPES

SUMMARY

The discussion of the motion of ballistic missiles would be inconmplete wthout some comment on the "work horse" of all trajectory analysts - trajectory simulations. It is beyond the scope of this report to discuss any simulation in detail. However, for the uninitiated, a few brief comments are presented regarding the basic ingredients of all simulations, the relative operational cost of several types of simulations, and some characteristics of a sixdegree-of-freedom simulation currently in use at APL.

'.

Pp

a
-325-

1.Preceding pa8~a blank

1
... .. .. .
-......
e.~---

,I,

I
THE JOHN4 HOPKINS UINIVERSIT

APPLIED PHY6SCS LABORATORY


SILVIO SPOINg MARtLAND

A trajectory simulation is a computer program by which the time history of the vehicle motion may be determined. Given a set of initial conditions and the particular quantities for which a time history is desired (outputs), the reentry body simulation consists of five major parts (see sketch, page 328): 1. 2. 3. 4. 5. Equations of motion, Earth characteristics, Atmospheric characteristics, Mass characteristics, Aerodynamic characteristics.

Each of the major parts may be very simple or very complex. The equations given in Section 4, 3 provide the basis for a very simple simulation in which the equations of motion are reduced to a single degree of freedom (motion along a constant flight path angle), the earth is flat and nonrotating, the acceleration of gravity is zero, the atmosphere obeys the hydrostatic equation, the wind is zero, and the vehicle mass and aerodynamic characteristics are defined by a single parameter (0). However, the usefulness of this simulation is limited to those studies for which approximate time histories of a very few trajectory parameters are sufficient. For many types of work this simple simulation is not adequate. Increasing the complexity (and cost) over the one-degree-of-freedom (I DOF) problem, a threedegree-of-freedom (3DOF) trajectory simulation may be used. Usually the 3DOF designation refers to degrees of freedom in the three orthogonal linear dimensions. In this case the simulation is also called a "point mass" simulation. The only mass and aerodynamic characteristics used are the body weight and axial force characteristics. Any atmosphere and earth model may be used. This simulation is reasonably uncomplicated but has sufficient accuracy for most types of flight test analyses, for defining targeting data, and for evaluating most types of range partials (sensitivity of impact dispersion to various anomalies).
-

327

Preceding page blank


................................,,..".,..

TH

JOHNS H4OPKINS UNIVKRIIII"Y

APPLIED PH.YSICS LABORATORY

LJ

LU

CC

u. U. L!J
0 . I. F

INI

cii

Il

32o

T141, JOHN@ HOPKINS LNIV|IftIY

AFPPL.IED PHYSICS LABORATORY


dILVO' SPRING MARYLANO

For evaluating the detailed dynamic behavior of missile motion and for calculating precise impact locations, a six-degree-of-freedom (6DOF, three linear and three rotational) program is required. In this simulation, the body mass characteristics are defined by weight, center of gravity location (specified by three coordinates), moments of inertia, and products of inertia. The aerodynamic terms include the axial force, normal force, center of pressure location, pitch (or yaw) damping, roll damping, and asymmetric force and moment terms as required for a specific study. This type of simulation is the only one (of the ones discussed) that accurately accounts for angle of attack effects, Any atmosphere and earth model may be used. There are many modifications to these thrbe basic simulations. For a given study the analyst will select the type of simulation based on a trade-off between accuracy and cost. For a typical reentry trajectory, the approximate machine cost of running the three simulations discussed above is as follows: lDOF 3DOF 6DOF Negligible, $1 per trajectory, $30 per trajectory.

As an illustration of a typical 6DOF reentry body simulation, one of the simulations currently in use at A1'L is described briefly. 6.1 Initial Conditions I. 2. 3. 4. 5. Geometric altitude, Missile velocity and flight path angle relative to rotating earth, and 0 components of (see Fig. 4-11), Initial roll orientation of the body axes with respect to the inertial axes, Body rates (roll, pitch, yaw). -329-

.. .

. . .. . . . .

?He

JOHN$ HOPKINS

UNIVt61UIT

APPLIED PHYSICS LABORATORY


616-IPI 6PiING MAILAND

6.2

Missile Mass Characteristics


I.

Weight, X, Y, Z coordinates of the center of gravity, Moments of inertia (about X, Y, Z axes), Products of inertia (about X, Y, Z axes).

2. 3. 4.

Provision is made for two sets of characteristics so that the trajectory for a two-stage reentry body may be computed. Provision Is also made for varying the weight and the two lateral coordinates of the center of gravity as a function of altitude, This option is used for studies In which body shape changes as a result of aerodynamic heating. 6. 3 Aerodynamic Characteristics

* :

Some of the aerodynamic data are in an aerodynamic subroutine of the program and some are inputs, The subroutine provides the axial force coefficient, normal force coefficient, center of pressure location, pitch (or yaw) moment damping coefficient, and roll moment damping coefficient. The axial force is a function of Mach number, angle of attack, altitude, and velocity. The center of pressure, pitch damping, and normal force data are functions of Mach number and angle of attack. The roll damping coefficient is a function of velocity and altitude. Provision is made for using two sets of data corresponding to the two sets of mass data. The aerodynamic inputs include the asymmetric pitching, yawing, and rolling moment coefficients as a function of altitude and multiplying factors as n function of altitude for varying the aerodynamic data included in the aerodynamic subroutine.

-330-

I
?MK JOIINO ,4OPW0i, UNtVftSINIY

APPLIAD PHYIDICS LABORATORY

0. 4 Atmosphere CharacteristicB into the The 1962 Standard Atmosphere is built temperature, simulation but any variation of density, may be used. wind, and wind direction with altitude 6. 5 Earth Model model is The WOS/60 ellipsoidal rotating earth used, 6.6 Equations of .Moton

axes) used The six equations of motion (in body given in Elqs. freedom are the three rotational degrees of linear dethree the (4. 5-1), (4. 5-2), and (4. 5-3), and grees of freedom given in Eqs. (4, 4-4), (4, 4-5), and and moments are com(4. 4-6), The aerodynamic forces are then puted in the body axis system. Accelerations transand velocities computed in the inertial axes, and lations are computed by Integration.

6.7 Output
The outVarious output quantities are available. fcelowing the provides put format currently programmed time histories: 1. (body axes), Three components of acceleration Total angle of attack,

2. 3.
4. 5. 6, 7. B. 9.

Pitch, yaw, and roll rates (body axes),


Mach number, Velocity (relative to rotating earth), Dynamic pressure, earth), Flight path angle (relative to rotating Altitude (geometric), of the Downrange and crossrange coordinates trajectory. -331 -

TM JOHNS HOPKINS IJNIVNISIY

APPLIrF

PHYSICS LABORATORY

SILVIO GPOIrN MARYLAND

The fiaal latitude and longitude are also listed. Several options are available regarding the printout time Interval. The integration time step is variable (with a minimum of 0. 0001 second), and readouts may be obtained at every time step, any multiple of the time step, every second, or any multiple of a second. For a 3DOF simulation only the first and second initial con ',,a, the first mass characteristic, the axial force coe at (as a function of Mach number and altitude only), and the three linear degrees of freedom are used, Hence the computer cost is lowered by an order of magnitude compared with the cost of using the 6DOF program,
'.,

J
i3
vi I.,
I,

...........................................................-

UNIvI" Oh s iasl IN$tlwt Amu LABORATORY PHYSICS APPLI-lD

IOV,2v
NOW,

MAmYAN0_

7. FLIOHT DYNAMICS DATA ANALYSIS

.'.

-333-

APPLIED PHYSICS LABORATORY


GILYIR1SPRING MAUYLANt)

?He

JOHNS

HOPKINS

UNIVRPSIT

SYMBOLS FOR SECTION 7 Symbol A


A a

IDefinition axial force


amplification factor semimajor axis of the dispersion ellipse

TypicaltUnits pounds

feet feet
-

b
iA

semiminor axis of the dispersion


ellLpse axial force coefficient, A/qS normal force coefficient, N/qS slope of the normal force coefficient, BC N /60 pitch damping coefficient, 2Cm/6(qd/V) per radianA
-

CA C N4 C

&

Cm
II,

q
D F g damping parameter (see Fig. 7-1 1) resultant transverse force resultant transverse acceleration,
-

pounds

2
g g g feet

longitudinal acceleration gy gz h * Iy K lateral acceleration normal acceleration altitude roll inertia pitch inertia parameter defined in Eq. (4. 3-6)

-t2 slug-ft
slug-ft 2

Preceding page blank ti~.~rniiI

-3

Tht JOMNI MOPKINI UNIVtltUITY

tPLI ED PHYSICS LABORATORY


ILV14 SPRIING M&mLANO

Symbol m N P p r q or q S V W XI YoZ

Def nition mass normal forcc atmospheric pressuro


roll rate

T slugs

lUnits

pounds lb/ft dog/. ec or ,rad/sec lb/ft ft2 ft/sec pounds

pitch rate yaw rate dynamic pressure reference area velocity we Ight body coordinates with origin at center of gravity. X coordinate Is parallel to the body centerline, positive forward; Y coordinate is perpendicular to body centerline, positive to the right looking forward; Z coordinate? forms a right-hand system with X and Y, positivo direction is down when looking forward, X X .distance cPpressure distance from nose to center of gravity from nose to center of

feet feet

angle of attack in X-Z plane angle of yaw in X-Y plane ballistic coefficient, W/CAS total trim angle of attack

degrees degrees lb/ft

degrees degrees feet degrees

v
&X Ai

flight path angle X c.p. c~g. change in trim maneuver plane due to roll rate resonance index (see Fig. 7-11)
-

336

APPLIED PHYSICs LABORATORY


MA 116Vfl BPONING. ,NVLAPID

Symbol

Definition orientation of trim maneuver plane, positive measured clockwise from Y axis looking forward

Typical Units

degrees

WA

Fig. 7-40) deg/sec aerodynamic frequency (see

Subscripts
c, p.

center of pressure center of gravity conditions at beginning of reentry (400 000 feet) conditions when p - 0 component in the X, Y, or Z direction, respectively

co g, E o X, Y, z

with respect to time. A dot over a symbol means the first derivative to time. respect Two dots mean the second derivative with

-337-

II
THIN IOINO HOPKINS 610NIV|nsIty

APPLIED PHYSICS LABORATORY


SlIV IA PIM* MARYLAND

SUJMMARY

A detailed discussion is presented ot the pocv(lures currently used at APL in the analysis of vehiclei.

motion during reentry, ments on: 1, 2. 3. 4,

This discussion includes com-

The basic data available, Basic data processing, Conversion to engineering quantities, Special analyses for dispersion and "roll lock-in.

Examples are given of: 1. 2, 3, 4, Typical flight test records, A method of extracting range time from these records, Typical body rate and acceleration traccs, Work sheets for computing angle of attack, aerodynamic pitch frequency, and several quantities related to roll resonance.

339'

Preceding page blank

7. 1 A vailiblo Dn2t:1

t1

:I In II

,1;I "trla~ og t dli ceeJ n- i tl r'I -,


lalu 111;11y,~z

iI I f(;r, ti vrst wdo commonoly lv avlibe 1'iio 1q111ht J(I CFInaL ,vsis am( tleC p itch, yaw\, :11]L1 roll -: tc .111 nl the

ctltvy dI spu e

is

-d

i.It:i Iiflud1 (
Xf

.i t: 1111. tt 0-

sphe-ric ptcsure and temperature-, nr' the wind speed


and di rection a s a fund; ion of a-Iltitudf,2. c1181'actCI'i stici ol' the rceentry body al e

tho mass

T hes e include the wei;lit., center of grevity 10o!Iatlon (three coor-I di ca te s , and tho ma rients aind r producets c nf Ui!ti1 T Ti!'
iriti;it eudIti.ons of. the roe nti'% tr-,Jfjocorv ay c21 I120 also ru-

viui(1

q.1 ir V !(.

These cond ions are fruequentl~y speciftled at

400 000 feet ,ititude, which is an altitude wvell abovel thle beng~ibc atilosphere, and inclucle the velouity, flighit Path rate. Additional initial conditions must be, specified if the, refntry tb..'y motion at very high riltituatis 1,; being e~fxamined or 1.f the precise impact, point must bke deter-. 1rrined. Thece quantities include the initial angle of a-ttack, side slip aniglu, toll ang. , pitch rate, ancl yaw rate..1 Although in generai ail or thest, data tire required fox a precise ainalysis of' the reentr~y d,,rn:inics, it ocCP.sionalty happens that some part of the danta is nct availabLlec. When this situation arises, the anatysi s somietinmes m~ay be co.-tinued without seriously compromising its accL-acy by substituting nominal values for the mitssing .

data.

As a`-iL npe the weather data may be unava.ila-I ble, but it may be possible to substitute weather data, (except for winds) that were, determined during prevituus tests in thait part of the world (and during that tim(! of the year) and still obtain satisfactory results, Or it may beI possible to ecoose one of the availabl~e standard atmospheres and obtain a. satisfactory result. Should thm acceleration at the reentry body center of gravity be unavailable, accelerations measured at other location.- may be used to calculate the acceleration at the

Precedfng page blank

-.

341-

TNH JOHN*I HOPKIIN UN""v6IY

APPLIED PHYSICS LABORAT'oRY


BILU

S1N PeiNG MASYCANO

center of gravity, An accelerometer pomitioncd nt n point away from the center of gra vity will bi influenced by the body ang-ular rates and accelerntions; the data from such an accelerometer mA'y be corrected to yield thc necelern.. tion nt the center of grnvitY by nieans of the following
eq uations:

"- measured g
mesrdg32X
Y measured g

3 z 2,-

(pq

"

P) +

-X

q 2+ r 22) (7-1)

2
Z 2",2327(rq + z
+
-

32.2
2 p+ )+" X
2

y 2 1 + 3---(r !) q2)2

(+pq)

(7-2)

(p'

S"Zmeasure + q 2) --

+ qr)

-2(pr-

!)(7-.:3)

7, 2

Data Sources

Redundant receiving stations are frequently used to ensure reception of telemetry from a reentry body. Ground stations, shipboard stations, and airborne stations may be used. The quality of the data received at any one of these stations may vary widely depending upon the geometry of the trajectory relative to the station location. As is now commonly known, a body reentering the atmosphere develops a plasma sheath that impedes RF transmission. The quality of the signal received through this sheath varies greatly with the aspect of the body relative to the station. Reentry bodies with a low ballistic coefficient (essentially a low weight to drag ratio) may have a plasma sheath of such impedance that transmission of the reentry data acquired during this "plasma blackout" must be delayed. It may be necc.;sary to examine the signal received by several stations in order to acquire the data needed for a reentry analysis. 7. 3 Data Format

Flight test data, as currently processed at APL, are in the form of either DITAR records or DITAP records. -342-

Ilk.

THR JOHN$ HOPKIPNSUNIVVISII' APPLIED PHYSICS LABORATC)RY SLuVER 5PRlN4 MARYLAEI)

In both cn ses the flight functions are digitized by dividing the calibrated range of the sensing instrument into 255 intervals. A value of the function may then be represented by a number of counts or bits ranging from 0 to 255. In the case of the DITAP records, the count level is plotted as a function of time, and the record is an analog display of the function, An example of the data format is shown in Fig. 7-1. Timing marks are applied to the DIITAP records according to the scheme In Fig. 7-2. In the case of the DITAR records, the bits are plotted In a condensed form that permits many telemetry functions to be displayed on one record. Two traces are required for each function, one trace to count from 0 to 15 and the second trace to count in increments of 16 from 0 to 240. By adding the counts from both traces, any number from 0 to 255 can be represented. An example of the data format for DITAR records is shown in Fig. 7-3. An example of the timing marks on the DITAR records is shown In Fig. 7-4. The series of marks on the margin of the record may be interpreted as time of day. The DITAR data format presents the data very compactly, and although a two point calibration is attached, it is sometimes difficult to establish the zero point from which one should begin counting the bits. From certain physical considerations, a value for one point on the flight re( ords can sometimes be established, and this point can be used to confirm that the correct zero bit level has been chosen, The following is an example of these techniques for confirming the. bit level. The pitch and yaw rate gyro data, before reentry, will have a sinusoidal shape. Except for the effect of principal axis misalignment, which is uuually small, the sine wave will be centered about zero rate. From the calibration tables used to convert to engineering units, the number of counts for zero rate can be determined, and this then fixes one point on the rate gyro data curve. Subsequent values can be determined simply by counting bits. Furthermore, the frequency of the pitch and yaw rate gyro sine wave is related
-343-

L:

Tlt JOHN* NOPKINI UNIVIE"I611

"

APPLIED PHYSICS LABORATORY

~~iiLV~g
*PSNQ

MARYfLAND

to the roll rate by means of Eq. (4. 4-21) when the body is unaffected by the atmosphere. Therefore the exoatmospheric roll rate can be fixed by determining the frequency of the pitch rate gyro output, and subsequent values can be determined by counting. The count level for the accelerations may be established by noting that the exoatmospheric accelerations are zero. 'The bit level, therefore, must be the one that corresponds to zero acceleration. Subsequent bit levels can be estnblished by counting. In the case of the longitudinal acceleration data, advantage can also be taken of the two ranges of Instruments that are frequently used. One accelLi'ometer is sensitive enough to detect the low level of acceleration at the beginning of reentry, and the second is sturdy enough to measure the peak g. The less sensitive instrument may be expected to go off scale at some point in the reentry. The count level is 255 at the point where the range of the instrument was just exceeded. The acceleration at this point may be determined easily from the calibration and, of course, the less sensitive instrument should also be recording this level of acceleration so that, by means of the calibration, the count level for the less sensitive Instrument is established, Future plans call foi a computerized conversion.':1 of the data from counts to engineering units as well as for machine plotting of the data in engineering units. The flight data most commonly used for dynamics analysis are the roll rate and the three accelerations. Typical examples of these functions (as well as the pitch rate) are shown in Figs. 7-5 to 7-8. 7.4 Data Processing
0

Further data processing depends upon a knowledge of the flight Mach number history. The flight Mach number could be calculated based on the longitudinal dec'.leration if the altitude were known; however, the altitude| history generally is not available. The recourse, then,

-344-

,i

:'

T"I JOINSt

WDOIKINl UNI""IvIt*TY MA O~ A AP D

APPL.IED PHYSICS LABORATORY


6i.l .W | U S PS .. .

is to determine the flight Mach number from a computer

simulation,

The results of the simulation must be checked

against the flight data to ensure an adequate simulation of the reentry trajectory, The easiest check that can be made is a comparison of the longitudinal acceleration history as predicted by the simulation with the flight data results. The velocity history of the body is obtained by integrating the acceleration history. If the simulated longitudinal acceleration history is verified by the flight data. and the transverse accelerations are small enough that the longitudinal acceleration represents the total acceleration, then the simulated velocity history (hence Mach number history) will be correct. For the purpose of matching the simulation results to the flight acceleration, it is convenient to compare the results based on the time after the occurrence of the peak g. The time from peak g to impact is a good check on the calculated Mach number history, If the acceleration from a nominal simulation does not agree with flight results, the factors in the flight that are not nominal (such as the aerodynamic asymmetries) must be included in the simulation in order to produce a match between the flight results and the simulation. Occasionally it is necessary to modify some of the aerodynamic coefficients in order to produce an acceptable match. At points where the altitude is known (generally the only point is at impact) the Mach number may be calculated based on the following equations. By definition of the axial force coefficient:

CAqS = A
or

(7-4)

C AS

(7-5)

.w

-345 -

.. .. .o

t!,I

tug JONN*

MOpKINS uNtIVURSTY

APPLIED PHYSICE LABOATCroY


lLVlftI 1PUIg N MARYLAND

We may introduce the detnition of dynnrmic pressure to give:

CtA (s/2 )PM tS W so that -

(7-()oM) gx

The axial force coefficient, CA,

is pri~ncipallyo a funotion

,I

of Mach number (and also a slowly varying function ofu anglt of attack and altitude). For a particular angle of attack and altitude, we can Plot CAdM2 versus Mach hsumbef eased on predicted aerodynamic coefficientsh We may then use flight data to calculate a value of CA M2 from Eq. (7-7), enter a. plot of CA M2 versus Mach number, and determine the Mach num ber. As mentioned previously, thixs se-heme can usually be applied only at the impact point since the altitude (which determines P and hns some effect on CA) is not known at other points along the trajectory. With the Mach number history established by matching the simulation longitudinal acceleration to the flight results, the angle of attar-k may be computed as follows. The maneuver acceleration is related to the transverse force by Newton's second law: SF 2 2 W = whore g Fgz z + gy

(7-s)

Expressing the transverse force in terms of the normal force coefficient, we have: CN rqS
-

(7-9)

346
I.

I
tHI JONN$ HOPKINS UNIVRlOrIY

jBfl

APPLIED PHYSICS LABORATORY


BL m ,~ Ui~lk AAL&kO

.1

From a similar development considering the longitudinal acceleration we have:

CA' CA

'gx'

(7-10)

The above equations are combined to give:

1
"gX CN The use of Eq, (7-11) with the flight values for the accelerations and the established (wind tunnel test, etc. values for the aerodynamic coefficients permits determination of trim angle of attack as a function of time, For routine flight test analysis, the envelopes of the normal and lateral accelerations are usually all that is extracted from the flight recorr's, and the mean value of the envelope at a particular point in time is used to calculate the trim angle of attack, A sample work sheet that shows a convenient technique for calculating trim angle of attack is shown in Fig. 7-9. An alternate technique for computing angle of attack is based on the rate gyro data. For small angles of attack: r + P (7-12)

q
o-

qI
f+

(7-13)

(7-14)

For the times of interest, & and are frequently negligible so that ai and 0 may be obtained directly from rate gyro data. The significant advantage of this technique -that neither the Mach number history nor the

347
S..
... . _ -. . . . . - . . . . . . ... _-.*. I ....... I I. i ." I ...

"

T~~~YI JOkN4S M'OIINIll LJW VRIllltT

APPLIED PHYSICS LABORATORY


6.60" SPICING MANYLANO

aerodynamic coefficients must be known - is obvious.


The drawback is the required accuracy of the rate gyro data, Rate gyros hnve a tendency to drift so that the rates indicated may be somewhat in error. At low angles of attack in particular this bias may introduce a considerable error' into results obtained with the above equi.tion,

The trim i mneuver plane (i. u, the plnne of the transverse force acting on the budy) is simply calculated from:
1tan (7-15)

The aerodynamic frequency is computed from Eq,

(4, 4-36): N WA ;
y rad/sec

(7-16) (
We may re-

Note that CN. must be in "per radian" units.

write this expression in terms of the flight data by substituting for qS from Eq, (7-10):

g ,,N &,x
A

CA

Iy

' w

(7-17)

Flight values of gX and the established values for the aerodynamic coefficients permit calculation of the aerodynamic froquency. A sample work sheet which describes a conv(,nient format for calculating WA is shown in Fig, 7-10, Depending upon the closeness of the aerodynamic frequency to the roll rate, the trim angle of attack may be amplified to a value different from the result that would be attained by a nonrolling body, as has been discussed in

348

I.

?we5JOHN$ MoPIIMIW UNIVMIitY A'APPLIED P'HYSICS LABORATORY

Section 4. 5. For some purpouses in reentry flight analysis, it is convenient to have a knowledge of the amplification factor. A sample work sheet that indicates the data necessary to calculate the amplification factor is shown in Fig, 7-11. The ratio of roll rate to aerodynamic frequency also affects the direction of the trim transverse force. A technique for calculating the change in direction of this force A6 is described in Fig. 7-12, 7. 5 Data Analysis A convenient technique for demonstrating the closeness of the reentry body to resonance is to plot both the roll rate and the aerodynamic frequency as a function of time on the same graph. As discussed tn Section 4, 5, when the roll rate is close to the aerodynamic frequency, the body is near resonance. Near resonance the previously mentioned amplification factors may be large, and large values of trim angle of attack may develop. It is possible that the large transverse forces associated with resonance (in combination with a center of pressure/center of gravity offset) may cause the roll rate to follow the aero>idynamic frequency. This effect, known as "roll lock-in, " may be easily detected from the plot of roll rate and aerodynamic frequency versus time. Roll lock-in is usually considered to be an undosirable phenomenon because of the large lateral loada, localized heating, and high roll rate associated with it (which may be sufficient to cause structural damage and which will cause some impact dispersion due to the abnormally high drag encountered). Of course, the roll rate may vary so that the rate reaches zero (a condition that usually leads to large impact dispersion), and the aforementioned plot of roll rate versus aerodynamic frequency displays this condition also. A te,'hnique for estimating the dispersion that results when a reentry body spins through zero roll rate .

346)

-4k

?M

JONNS

HO*MIIII4gUNIvImsltv

LABORATORY APPLIED PHISP7. BtLvlS 5,aNQ MANrfl,,n

is described in Ref. 7-1.

The semimajor aixi.s of the dis-

persion ellipse can be estimated from:

--

sin 512 YO
V

"i' v,

feet fu

(7-1B)

rIP

and the semirninor axis is: b = -a sin*/ Yo, hn feet.

If a simulntion for the basic tnajeetory is available, nd Vo aRV obtained for the trime when the flight

record indicates zero roll rate, (However, it usually is not necessary to simulate the roll or, transversc accelera-

tion behivior in this simulation. ) Values of ;o and Po are


obtained directly from the telemetry records, If a simulation is not availablc, 'YO, ho and Vo may be estimated as follows: 1. Y Assum eV
P

2.

Compute K o

-P 0 Y /sin

at the time ofn p

using Eq. (4.3-22) aid the gx telemetry trace, 3, or, thus hIc 0f 4. Compute V0 using Eq. (4.3-5). VE CASY compute P and

350 -

U
I

?41 .OM~S HSOiqNS

UNgvI~mUIV

APPLIED PHYSIC3 LABORATORY


616.VRPl $PRIMA MI&ykNV6AN

REFERENCE

7-1

B. F. Fuess, "Impact Dispersion Due to Mass and Aerodynamic Asymmetries," J. Spacecraft and Rockets, Vol. 2, No. 10, October 1967, pp. 14021403.

-351,I

"*

...

~~~~~~--

-m

ll'T14 JOMINl MDPkIN4b UNIVERI1~il

AIOPL ED PHY SICS LABORATORY

In

IItUt,
It,,

66

I,, ""IIIQ

cc

z c D:z

m Du

SU.
"7

-aL Preceding~ paebln

Preceding page blank

35

?MR6 JOHN$ HOPKtINS

UNIVtINIIy

APPLIED PH'YSICS L.ABOrATcORY

0Jw

La CC, l~

LL.

V) Ll.
-~ S

2D

APPLIV)P YS.CS LABORATORY

SIVOSRNGMOiR

Call

n41
m
tIo.
ml

Wri5

w
U.

'-I
~tI

)
.. j~z

y
w

--I

a,

U.

355-

APE~ LIED PHYIIC,5 LAPORATORY

0on-

0o -j
0-1

r4

0-

a2)

WW-'N MV

8z

1Lu w
2

<
U

I-'ec

z.
-

..

U j
'U

-36

I
?NI J0H.I

APPLIED

PiYSICS
IPU..G

MOPWINS UNIVgNhi1 LAOINIORV


MAELANO

I,vum

'I

11
F w -J
a

AZ
r4 U-

(pUOKh3IS.h511.LVII

1iOh

I
II

............

APPLIED) PHY'51CS L ABORATOPRY S.L~lm SPRINGMARThAND

U" IvCNsIY tI.t JOHNS HOPKINS

wj

I I/I

I'J

0
cc-

Ii
ID

sz
L

16)N~lVU3300

-58

TH9 JOHNS NoMIRIM UNIVXRSITY

AP'PLIED PHYSICS LABORAkCRY


NILVERSPRINIG MARVLAND

,
uJI

9
wh

!I.

(5) N0I.LVW313D0ViviuflINDO-1

-359-

YI4K JOHN* HOPKINS uNiVEP~iyI APPLIED PH*YSICS LABORATORY

S~vlmSPURS

M~~hAA

44

TH| JOHNS HOppfINImYNIvllhI1y

APPLIED PHYSICS LABORATORY


SIM- IRR-ND MA11LAND

0010
t-

IMPROVED VALUE FOR ANGLE OF ATTACK

NEW VALUE FOR NORMAL FORCE COEFFICIENT SLOPE, REQUIRED


IF ESTIMATE OF' ANGLE CF ATTACK IN

WAS NOTACCURAIE

GO
z
u

ANGLE OF ATTACK IDEGREESI


NORMAL FORCE COEFFICIENT SLOPE FROM LISTING OF AERO DYNAMIC CHARACTER. ISIICS IPER DEGREE) ESTIMATE OF ANGLE OF ATTACK REQUIRED

'

"2

NORMAL FORCE COEFFICIENT CALCULATED FROM FLIGHT DAIA IDIMENSIONLESS)

AXIAL FORCE COEFFICIENT FROM LISTING OF AERODYNAMIC CHARACTERISTICS

G u
'QB'rAINtD
z

(DIMENSION LESS)"']

FROM COMPUTER SIMULATION. USE TIME OF PEAK LONGITUDINAL DECELERATION TO MATCH FLIGHT DATA TO SIMULATION (DIMENSIONLESS)

Q00100
LONGITUDINAL DECELERATION (g)

NORMAL ACCELERATION 1g)

LATERAL ACCELERATION (g)

01 3 !

,,U)

Fig. 7-9

ANGLE OF ATTACK WORK SHEET

361 -

.i
-...... .. ... . .. . . . . . . . , . . . , -,

TI JOHNS HOPKINiS UNIP tlIo'Y A4PLIED PH-YSICS LABORATORY


SILV 2P00tkt M*EVLANL

@x@ W'A
@' K

AEPODYNA.-.,,:' f REQUENCY WiE G It.E ,' fCONDI

K'.

3 J/,,W

01

8 x(
Xc~g'
-

CENTER OF GRAVITY OBTAINED FROM PREFLIGHT DATA (INCHES) OF PRESSURE OBTAINED FROM LISTING OF AERODYNAMIC

SCENTER
p.

CHARACTERISTICS (INCHES)

(D xG

_____

CA __ CNt't -ox MACH NUMBER TIME BEFORE IMPACT

AXIAL FORCE COEFFICIENT OBTAINED FROM LISTING OF AERODYNAMIC CHARACTERISTICS (DIMENSIONLESS) OF NORMAL FORCE COEFFICIENT SLOPE OBTAINED FROM LISTING AERODYNAMIC CHARACTERISTICS (PER DEGREE) FLIGHT TEST, LONGITUDINAL DECELERATION (gI) OBTAINED FROM COMPUTER SIMULATION. USE TIME OF PEAK LONGITUDINAL DECELERATION TO MATCH FLIGHT DATA 1O SIMULATION (DIMENSIONLESS)

SCA

WA

57.3.A__Cl

(XI.

h AX)x

Fig.7.10

AERODYNAMIC PITCH FREQUENCY (WEATHERCOCK FREQUENCY) WORK SHEET

362

T"N%
UILYS,. I~iN

JOHNS HOPKINS UNIV~lItIIl?"

APPLIED PHYSICS LABORATORY


MAmvtANnI

AMPLIFICATION FACTOR

14

CMCI

PCOBTAINED

CHARACTERISTICS (DIMNOEIGH LdAT) (

PIEVIOUFROM

OALCULATIO

CH@CTMSTC IDMESINLSS
CN t R LOPE OBTAINED FROM LISTING OF AERODYNAMIC T OBTMAL FORCE COEFFICIENT CHARACTERIOMTICO PERRAILAN)

x N"JMU. A SC

II

PITC DAPOLINGCEFFIACAD9IPITC FROM IBANED RTMISTN FAROlAI D) DATA ( FLIGHT ARCTION DECELE LONGITUDINAL FROM LIS'TING' O'F 'AERODYNA M'IC NDINION~iBTAINED FORCE ITAXIAL OE r F ICI'EN CH RAT

l-

F@AF.RODYNAMIC PRQUNC

OBTAINED FROM PREVIOUS CALCULATION,:

" 4

2x Iy (PITCH INERTIA)m Ix (POLL INERTIA! AND OBTAINED FROM PREFLIGHT DATA (degre-/-econd)-

FLIGHT DATA (degreeCsiscond) ROLL HATE VELOCITY OBTAINED F'ROM COMPUTER SIMULATION (fwlsa/conll) F~REQECJ C, NUMBE.R TIME BEFORE IMPACT,,

SMACH

TIME OF PEAK LONGITUDINAL COMPUTERLSIMULATION, FROM TO' ObTAINED DECELERATION MATCH FLIGHT DATA TO SIMULATION IDIMSNSIONLES-S)

__

. I C(.

.. IX . . ........ C,, qd1 ly-1

t,

A-

p2 --- 2 I1-.y-)

I D.

xWJ VCA ...

..

F ig. 7.11

AMPLIFICATION FACTOR WORK SHEET3

.,

- 363 -.

TMI JOHNS HOPAINS LPlVntII i. {

APPLIED PHYSICS LAeORATORY


SIL*V|ISPII~IGN
hI*IyLAPIO

Vi.

7..i

AlP

SET SIGN OF AV/iTO Eli OPPOSITE TO THAT OP THE RO0LL RATE

!I

SIN A
A W 0.

0CL 0bi
61)

03 O TI ME
IMPACT

(COL. GJ FIG.7,11T
(CUL1

3 1N Al
Fig.7-12

MA

ROTATION OF MANEUVER PLANE WOPK SHEET

"

,laj,.

--

. .

-364-

two aw

" NOaImON w vNUMIV

APPLIED PHYlICB LABORATORY


$It.v1m EMPIN. MoYlvaNI

SS.

CONSTANTS AND CONVERSION EQUATIONS


K

"1

ii.

MIL
"U
-

205

,I ,

, ,

?TH4 JOHN$

HPKINS

UNIVSINYV

APPLIED PHYSICS LABO9RATORY


SILkVil 1 kipla MAVL.&h

Conversion Equations
Degrees Centigrade 5 ( degrees Fahrenheit
-

32)

Degrees Rankine - degrees Fahrenheit + 460


Degrees, Kelvin = degrees Centigrade + 273 Time (In hours) at zero meridian - local time (in (in degrees) hours) + local longitude 15 where West longitude is positive East longitude is negative Conversion Constants 1 radi-h 1 meter 57, 30 3. 28 feet

1 nmt " 6080 feet 1 gram/cubic centimeter 1 kilogram/cubic meter 1 millibar


=
*

1, 94 slugs/it3
3 1.94 x 10-3 slugs/ftt

2, 09 lb/ft2

1 knot

1, 69 ft/sec

36

Preceding page blank

APPLIED I'HYRICS LABORATORY


KrIL-11 NOMING M&ANVAND

TNI JONNs "OPKINS UsNIVNIIrTV

L
2
g :32. 17 ft/sec
2

Constants
Acceleration of gravity at sea levl Ratio of specific heats 1. 4 ft2 sec 2 degrees

j.

GCas constant for dry air = 171 Rankine

Gas constant for water vapor - 2759 ft /sec

degrees

Rankine
Radius of earth = R = 3441 nml we 2.09 x 107 feet 0. 729 x 104 rad/sec

Rotational velocity of earth NapiterLan base e


=

-4

2. 718

i.

Itt

.I

V!
-368-.

I
i

II

APPLIED PHIYSICS LAUONArORY


Sesyll

M "osUN T
VQIWS MAftT6AND

INDEiX

36

THU APPLIED PHYSICS 1.ABORATORY SPAIN& MANUAND.

JOH 911d

H PLIM6t ~NS

URSYIATV

~SILVIR

Accele'ration, Lateral
a convergence
-

-rolling

nonrolling missile missile

152 154

Aerodynamic Coefficients Asymmetric moment for flap on a cone Asymmetric moment for a. distorted cone Definitions Equations for a cone at high Mach numbers

28 29 10, 25 23 24 13 21 21 1 25 15 10 147 53 53 171 181

Aerodynamic Forces
A symmetric Axial Drag Lift Normal Aerodynamic Moments Pitch Pitch damping Roll damping Aerodynamic Pitch Frequency Altitude Geometric Geopotential Amplification Factor Angle of attack Roll r-tcceleration

Preceding page blank

-371-.i

TH| JOPNI HOrKING UNIVK'ICIA

APPLIED PHYSICS LASORATORY &ILANfl WILVIW*P.. AIN

Angle of Attack Amplification factor Convergence - nonrolling vehicle


- rolling vehicle

171 148
1 51

Definition Trim Angular Momentum, Conservation of Apogee


A symmetry

1 3, 155 26 81 86
!

Center of gravity or center of pressure off set EWect on roll rate In plane Moment (couple) Moments of inertia, unequal Out of plane Produ,-ts of inertia Types Atmosphere Exponential Polar Standard Tropical Standard 1962 Standard Axial Force Azimuth, Inertial B Ballistic Coefficient Ballistic Density Ballistic Wind Barosphere -572117 285 283 49 4 174 180 177 168 217 1*77 207 167 A 55 66 66 68, 69 13 103

tH I
! 1APPLIED

JOHNS

14OPKING

UNIVEEU2TV

me. MAN.*Sb SInSovmIP

PHYSICS L 'BORATORY

Boa5, Rate Effect of product of inertia in vacuum Transverse Effect of unequal pitch and yaw moment of inertia Boundary Layer C Center of Pressure

208 140 217 18

13

Cone
Aerodynamic coefficients Space Constants and Conversion Equations Continuum Flow Coriolis Force D Damping Pitch Roll Deceleration, Reentry History Density Ballistic Polar Standard Atmosphere Range partial for Tropical Standard Atmosphere Weighting factor for 1962 Standard Atmosphere 285 66 285 66 2B5 68, 69 23, 24, 29 140 367 20, 260 59

15, 24 16, 24 119, 120

'

373

1' -~..
-- . . . . . . . . . . . . . . . .. . '. " :'.. . . . .,,: *-=-_

-. -

THi

JOMNlI HIPKNS UNIVIIIIT"


SIVI I~)N M&VALND

APPLIED PHYSICS LABORATORY

Dispersion Axial force coefficient errors Density deviation Missile weight error Trim angle of attack - rolling missile - nonrolling missile Wind Drag Dynamic Pressure Definition Reentry history Dynamic Stability E Energy Equation, Conservation of Equations of Motion Six-degree -of-free-dom - symmetrical missile - asymmetrical missile Exosphere

286 285 286 288 289 283 21

17

119 146

81

137 168 49

Flight Path Angle Flight Path Angle, Inertial Flight Test Analysis Flight Time Ballistic trajectory in vacuum Reentry history

81 103 341

84 116

Free-Molecular Flow

20, 260

-374-

;L

.. .

.............

APPLIED PHYSICS LABORATORY

?He

JOH4NI HOPKINS UNIV9I0IYT

Frequency Aerodynamic or weathercock Body pitch - in vacuum G Gravity Effect on ballisti.c trajectory Law of Variation with altitude H Heat Absorbed at stagnation point Rate at stagnation point Homosphere Humidity, Effect or. Density Hydrostatic Equilibrium Equation
I

147 139

81 81 68

119 119, 120 50 56, 67 51


L

Inertia

Moment of Product of

217 207 177 59 K

In-Plane Asymmetry Isobar

"KnudsenNumber
L Lateral Acceleration

19

152, 154

-375 -

Tlt JOHN* MOPfINN U%4va*I1Vlt

A&PLIED PHYSICS LABORATO-PY


6LV6Rm100106 MA**VLAND

Lock In
Condttlons required for Definition Effect of roll damping Effect of roll torque 188 186 198 198

Longitudinal Deceleration M Mach Number


;

119, 120

17
68 65 65 83 56, 67

Mean Free Path, 1962 Standard Atmosphere Mesopause Mesosphere Minimum Energy Trajectory Moisture, Effect on Density Moment Asymmetric Pitch Pitch damping Roll damping Moment of Inertia, Unequal Momentum, Conservation of Angular N Normal Force

28, 29 25 15 16 21 7 81

13

Perfect Gas Law

52

-376-

!V.

APPLIEO PHYSICS LABORATORY


SIllVIP SPRING. MAtfLAND

?MR JuINIG

HOPKINS UNIVgRIgTlt?

Pitch Frequency, Aerodynamic Polar Standard Day Pressure, Dynamic Pressure Force, Horizontal Pressure, 1962 Standard Day Product of Inertia Criterion for lock in

147 66 17 59 68

215

Definition
Effect on body rate history, vacuum Effect on trim angle of attack
R

207
208 211

Radiosonde Observation (RAOB)

55

Range Partial
Density

285

Wind Range Sensitivity Factors Reentry Dynamic pressure Heat absorbed Heating rate Longitudinal deceleration Reynolds number Time Velocity Resonance, Roll-Yaw

283 87, 93, 94 119, 119 119, 119, 119, 118 118 120 120 120 120

Definition
Effect on Effect on Effect on First and roll acceleration trim angle of attack windward meridian second

171
180 171 173 187

I'f

S-a377S.
.. . .. . .. .. . . .. . . . . .

T?4

JOHN* MOPrINS UNIVIrNiMY SLyER SPaig..


M&IyLA&ib

APPLIED PHYSICS LABORATORY

Reynolds Number Definition Reentry history Rocket Observation (ROCOB)


Roll Damping Roll Rate

18 119, 120 56 16

Effect of aerodynamic asymmetry Effect of center-of-gravity offset Effect of product of Inertia Effect of resonance Effect of unequal moments of inertia Equilibrlum

178 175 213 181 219 1 81

Scale Height

Definition
1962 Standard Atmosphere Second Resonance Simulation Slip Flow Space Cone Speed of Sound Definition Polar Atmosphere Tropical Atmosphere 1962 Standard Atmosphere Spin Up, Spin Down Standard Day Polar Tropical 1962

55
68 187 327 11, 140 6I1 66 66 68 187 260

66 6"68, 69

-378I '

tfm

JOHNS HOPKINS LMNiVCR1ITY


W6V~iN SPQ1%* MANT6ANAD

APPLIED PHYSICS 6AIORATORY

Static Margin Static Stability Stratopause


Stratosphere Sutherland s Viscosity Equation

14 14 65
65 35

T
Temperature

Polttr Standard
Standard Tropical 1962 Standard

66
66 6B, 69

Thermopause Thermosphere Time History, Reentry


Trajectory

65 65 118 83 327 140 66 65 65 V

Minimam Energy Simulation

Transverse Body Rate Tropical Day Tropopause Troposphere

Velocity, Missile Inertial Mirimum Energy Trajectory Reentry history


I

103 B3, 90 118

I"

I,
A'1

r379

UNtVIN5"YV THK.JOHNS ONpK"IN

APPLIED PHYSICS LACORATORY


*IhvtRl SP*lNm M&RYLVAIN

Viscosity Definition
Sutherland's equation 19[62 Standard Atmosphere W

18
35 68

Wind

Ballititc

283

Geostrophlc Range partial for Weighting factor for

60 283 282

380 -